DISCLAIMER: Vacuum tube circuits work with dangerously high voltages. Do not attempt to build circuits presented on this site if you do not have the required experience and skills to work with such voltages. I assume no responsibility whatsoever for any damage caused by the usage of my circuits.

All rights of photos and text reserved. Usage of photos or text from my blog on other websites or for any other purpose only with prior permission. If you want to use any material from my blog please contact me by email.

Thursday, January 30, 2014

Music: John Zorn, Spillane


I got John Zorn's album Spillane as a gift from a friend. This is a great album which I probably never would have bought myself and missed. Thanks a lot to Hartmut who gave this record to me many years ago. I still cherish it!

The album could be categorised as avant-garde. It was published 1987 on Nonesuch records. It is meant as a homage to crime novel author Mickey Spillane. The first side only has one long track which is a continuous collage of sounds voice and music sampled from a turntable, occasionally mixed with some instruments.

Snippets from movies, TV shows, a voice reading from a newspaper. These sometimes appear like an audio drama or a sound track and intend to give an impression of Spillane's main character, Mike Hammer. The second side begins with a title which Zorn wrote for guitarist Albert Collins. It opens with Collins' bluesy guitar play and could be mistaken for a blues title. But halfway through the track it changes considerably. First with a voice reciting some text. Then a groovy hammond organ joins and it appears like a completely improvised part.

The second and last title and side two is again a rather wild collage of sounds, string instruments and voice.

This record is definitely not to everybody's taste but worth checking out. The recording quality is exceptional. Very transparent and neutral. Some of the sounds pop frighteningly realistic out of the speakers. There is lot's of details to discover. A record which can be played over and over again to discover new things.

Best regards


Saturday, January 25, 2014

Tube of the Month : The 6X5


This month I'd like to present another small rectifier suitable for preamps. I have mostly been using the 6BY5 for that. Last year I introduced the 6AX5 which has the advantage of lower heater current. There is another rectifier which needs even less heater power, the 6X5.

With it's moderate 0.6A heater current this tube is perfect for use in preamps when space is restricted and only small power transformers can be used.

The tube has an octal base. The pinout is shown on the left. Like the 6AX5 it is a full wave rectifier with two plates and a single, common  cathode. It can deliver up to 70mA of DC current and up to 400V output voltage which should be good enough for most preamplifiers. See the data sheet for all technical parameters. The tube would be good to be used in my Octal line preamplifier or Octal phono preamplifier. Other suitable applications would be in power amps which use separate supplies for driver and output stage. The 6X5 could power the driver and/or input stages. If space is restricted the 6X4 could be used instead which is very similar but in a smaller envelope and on 7 pin miniature base.

The 6X5 does not seem to be used a lot by amplifier builders, hence availability of NOS tubes is good at low prices. The tube can be found from all the major brands of the vacuum tube era. Let's have a look at some of them.

RCA 6X5:

Cunningham, made by RCA:


Another Sylvania:

One more in the more modern Sylvania packaging:

This later tube has a different internal construction as the older ones. Here the two in comparison:

A militay tube, JAN (Joint Army Navy), also made by Sylvania:

Philco, probably also made by Sylvania:

Another Philco, this one looks as if it was made by RCA:

Tung-Sol 6X5GT:

General Electric:

The GEs are shorter and more compact than the others:

The construction styles (arrangement of the plates) varies considerably. This one has the two plates arranged at different heights:

And this one side by side:

In the second tube it almost looks like there is only a single plate, so let's open one up to see how it's done:

A close up shows that there are two separate plates:

The two plates removed:

Let's examine the other construction style:

A close up of heater and cathode connections:

Removing the upper plate:

The cathode with the heater wire partially pulled out:

Cathode and heater:

A great tube which deserves to be used much more!

I hope you enjoyed the first tube of the month presentation of this year.

Best regards


Wednesday, January 22, 2014

ShellacSavor, Part 1 : Introduction


After last years European Triode Festival, I stayed another day in Berlin. I visited my good friend Frank W. who introduced me to Shellac records. I never seriously listened to shellacs before so this was a first for me and it was a big surprise. Frank uses a Octal Preamplifier Mk1 and a 801A power amp which I built. His speakers are an enhanced version of the Rho from David Haigner. For Shellacs Frank uses a mono phono preamp.

Shellac recordings reproduced over a good system and with a suitable preamp which matches the recording EQ curves properly (more about this later) are very special. The reproduction has not much in common with todays expectations towards HiFi. No deep bass, no extended treble, not much resolution and horrible surface noise. But that midrange is pure magic. The emotional impact is simply overwhelming. Voices not only go under your skin, they touch your soul.

Being mono, Shellacs don't reproduce a soundstage between left and right, if played through a regular stereo system. But they can give a beautiful impression of depth and space which is more natural than the artificial space reproduced by many stereo records.

We spent the whole evening listening to a lot of Shellacs. Mostly Jazz records from the era between World War II and the introduction of vinyl. Frank characterised the sound of the Shellacs nicely in his  typical Berlin accent : 'dat jeht direkt auf de Pumpe, wa?'. Translated this means: 'this goes straight to the heart', but this does not get the twist of his Berlin slang, a more appropriate translation might be : 'this goes straight on da pump, huh?'.

I have heard and built a lot of systems and listend to different audio media during the last decades. It is difficult to impress me with something. But what Frank showed me opened a whole new world for me to explore. Of course this does not mean that I will switch to Shellac entirely. I will also not change the name of the blog to ShellacSavor. Vinyl will continue to be my main source and I am also not advert to digital. Shellac is a beautiful addition for a different kind of music reproduction. Thanks a lot to Frank for introducing me to the world of Shellac.

Frank had been asking me to build a dedicated phono preamp with some more EQ curves than the unit he has since a while. Now after I heard this it was clear for me that I have to design a proper phono stage with as many EQ settings as possible.

I already built a Phono preamp with variable EQ curves in 2011. But that preamp mostly only covered EQ settings for vinyl mono records, not for shellacs. So I started with researching the EQ settings used by various recording studios for their shellac pressings.

The basic shape of the equalisation needed is similar to that of vinyl records, it only varies in the corner frequencies. This is how a generic EQ in the reproduction looks like:

At a certain frequency, which defines the 'rumble shelf', the response starts to roll off. Attenuation then stops at the Turnover frequency and stays flat until the treble roll off point after which the response rolls off again. The Rumble shelf can extend up to 100Hz. For RIAA it is 50Hz. Some shellacs have been recorded without a rumble shelf. But still that frequency had to be set somewhere since the amplitude would otherwise rise to infinity when moving towards 0Hz. Typically that was around 15-20Hz.

Rumble shelf and Turnover frequency have usually been named by a 3 digit number, followed by a letter. The 3 digit number would indicate the turnover frequency. For different rumble shelfs letters have been assigned.

Some examples :

500R = 500Hz turnover, 50Hz rumble shelf (this is used for RIAA)
500N = 500Hz turnover, no rumble shelf (15Hz)
500C = 500Hz turnover, 100Hz rumble shelf
250N = 250Hz turnover, no rumble shelf

Most shellacs seem to have been recorded with no rumble shelf. I could find information about 4 rumble shelf settings used at the curing process:

N - None (15Hz)
R - 50Hz
B - 62,5Hz
C - 100Hz

I could find information about these turnover frequencies used : 200, 250, 300, 350, 400, 500, 630, 800.

Not each turnover frequency has been used with all the rumble shelf settings. In total I came across 12 combinations which have been used:

200N, 250N, 300N, 350N, 350C, 400N, 500N, 500R, 500B, 500C, 630N, 800N

It is apparent that there was a great variety in recording curves. Rumble shelf frequency spreading from 15Hz up to 100Hz and turnover frequencies spanning 2 Octaves from 200Hz up to 800Hz. That translates to differences of up two 12dB in the frequency region where the ear is the most sensitive. For serious shellac reproduction an adjustable equalisation is mandatory.

There was an equally wide spread in the high frequency EQ. This was not defined by the corner frequency at which the roll off starts but rather by the roll off in dB at 10kHz. I found indications of 12 different high frequency equalisation settings:

0dB (none, or flat), 5dB, 6dB, 8dB, 9dB, 10dB, 10.5dB, 12dB, 12.7dB, 13.7dB (same as RIAA), 16dB, 18dB.

Again a huge spread. A difference of 18dB at 10kHz for different records. Here it is even more apparent that a shellac would sound very wrong if the treble EQ is not done right. It could either be totally dull or annoyingly shrieking.

The EQ curve used for recording is specified by a combination of both, for example:

500R-13.7 (RIAA)
300N-16 (used by Columbia)
400N-0 (used by Telefunken)
500B-16 (NAB, NARTB)

For a more comprehensive list of EQ settings please see here.

Many different combinations had been in use. In order to implement a preamp which can reproduce all of them correctly, it makes sense to use two separate selector switches for rumble shelf/turnover and treble roll off. This way all the settings can be implemented with practical 12 position switches.
Of course such an implementation will allow combinations which had never been used, so the setting should be done according to a list as the one mentioned above. An advantage of such a dual switch set up is the possibility to also fine tune the sound and deliberately use a 'wrong' setting as a tone control. For example a roll off setting with less attenuation could be used to boost high frequencies a bit for dull sounding records or to attenuate it if it is too bright sounding.

Here is a sketch of the faceplate for the variable EQ preamplifier:

The circuit development and tube selection are underway. Stay tuned for updates about Shellac reproduction.

Best regards


Monday, January 20, 2014

Making of a 300B Amplifier, Part 1 : The Power Supply


As mentioned in my last post of 2013, I have a 300B amplifier in the making. I decided to build this amp in anticipation of the german made ELROG 300B which is in development. This amp will serve as a test vehicle to evaluate the tubes. The amp was actually scheduled to be built in December but got delayed a bit since I had to wait for some chassis material. In the meantime the materials arrived and while I wait for some stuff which is needed to continue with the 211/211 amps, I finished the single ended 300B mono blocks. Of course this amp deserves a 'Making Of' series of posts. I will start with the power supply.

The power supply concept follows my usual philosophy: Full wave bridge rectified B+ with 4 6AX4 TV dampers. Choke filtered (choke input) with an LCLC section in the PSU chassis, followed by individual LC decoupling segments in the amplifier chassis.

I prefer to heat the 300B filaments with the cleanest possible DC. While AC heating is ok for 45 or 2A3 amps, with the 300B I want DC for absence of any modulation of the signal by the filament supply. While current regulated filaments are quite popular and work well, I still prefer a purely passive approach with an LCL supply. LC in the PSU section followed by a local L in the amplifier.

Since the driver will also be a directly heated triode, two independent filament supplies are needed, which makes a total of 4 chokes in the PSU chassis. The amplifiers will be mono blocks, each with it's own external supply.

The chokes and filament transformer are mounted on sub assemblies which go inside the chassis:

The sub assemblies prewired and inserted into the wooden frame:

The high voltage power transformer and rectifier tubes will reside on top of the chassis along with the  B+ smoothing caps:

Pre wiring of the top metal plate as far as possible before it is placed on top of the chassis:

Doing the last connections before the lid is closed:

The finished power supply:

This time with a new color scheme. I found the metallic bordeaux red of the transformer cover to make a beautiful contrast with the metallic graphite grey of the capacitors:

The rectifier bridge:

Stay tuned for the assembly of the amplifier section and listening reports of this amp.

Best regards