RIAA-2010

Phono stage based on Transcendent RIAA
Copyright 2010 © Troels Gravesen

POWER SUPPLY AMPLIFYING STAGE CABINET SCHEMATICS MEASUREMENTS

Installing the SILK MC trafos for my Dynavector XX2 mkII cartridge made an external power supply necessary and the PSU section was placed in a separate cabinet making the whole thing quite bulky. Re-building the TRANSCENDENT phono stage has be on the to-do list for a long time and besides getting a smaller unit next to the grammophone, improving sound from hardwired amplifications stage, full PP capacitor power supply and better caps in riaa-equalisation step wouldn't hurt either.

While waiting for the right time to start building, a number of RIAAs have been auditioned and nothing came up that convinced me to go elsewhere. For the separate power supply I decided to make a circuit board as the number of components and overall complexity would make a hard-wired solution look like a bird's nest. No tubed regulation was decided either as I don't think it pays off. Re-building may not be the right word as all parts will be new including power supply, thus a semi-hard-wired copy.

Finishing the phonostage and setting it up for the first time is nerve wrecking! Will it be silent or take you into a new hum-war? I was exactly minded for another round of hum fighting, but this proved an unnecessary worry. Hooking it up and turning the line stage volume control max only left the usual sssshhhh of the universe. Yes, I do think it sounds a tad better but I don't blame you for thinking me biased from all the working hours.

Go to pics of finished RIAA.

Most parts were bought locally (ELFA, not cheap but around the corner), except for: Teflon sockets from Hong Kong (search eBay). MC trafos from SAC Thailand (read below). Tubes bought on eBay from Russia (search eBay). Hammond choke came from Finland (hard to find Hammond + decent shipping rates for a single unit).

The Power Supply Board
(BACK TO TOP)

Having a special transformer made with all secondary windings would be rather costly - and having a range of spare transformers - the solution below was decided. Included is a small 9V transformer allowing the power supply to be turned on/off from the RIAA box via a low-voltage DC line. Hunting for expensive Black Gate electrolytic caps is fortunately no longer needed and even for (valve) power amps, film caps seems a much better solution - if space allows, because film caps certainly are bigger than the old electrolytic caps. Here 22 uF/400V PP Cross Caps from Jantzen is used throughout.
A 470k resistor was added to HT supply below board. Turning off the RIAA leaves some 100 volts on the HT rail and my fingers just don't like it!

Finished PSU board. And it works :-)

The amplifying board with room for 2 x 22 uF/400 V polyprop decoupling caps.

The amplifying stage is going to be hardwired as circuit simplicity makes this easy. Tube teflon sockets have been bought as these are now available from Hong Kong at reasonable price. The quality of these teflon sockets is second to none.

Preliminary wiring. Click image to view large.
Let's see how things turn out with the soldering iron in hand.
The other channel is mirrored and obviously looks different.

Checking out wiring, correcting errors (...) - and no smoke or bad smells. Nice!

Bought a box of 6C45pi tubes from Russia - and it's nice to have a box of these tubes - from Russia. They cost very little, are as good as new ones from Sovtec - and for the input stage I found out I had to find matched pairs to get the voltage levels right. Finally I could manage close to 200 volts on V2 anode and 100 volt on V1. My Transcendent board can't make that - although it appears to have little sonic impact - if any.

Looking for more 5751 tubes, I came across the Russian 6N2P tube being a substitute for ECC83 - only having different filament supply. Having a handful of these tubes, I rewired the filament supply and added 100 ohms to ground from the connection between the two tubes. Pin nine was connected to ground. Preliminary listening tests suggested this was indeed a very good substitute and together with the 6C45pi tubes I now have an all Russian tube set-up; my Ruska-RIAA!

The coupling caps: Seen above is Audyn tin foil caps and these were later replaced Obbligato tin foil in teflon. Having done numerous experiments with tin foil caps from Audyn, Obbligato, Mundorf and Angela I don't find any significant different in sound quality. The much cheaper non-teflon caps from Angela, Audyn and Mundorf sound as good as the expensive Obbligatos. I used the Obbligatos because I had them.

Left: Close to completion. Right: Temporarily wired for MM only. No hum from running the Dynavector XX2 mkII directly into this.
Actually quite some level can be achieved even without the trafos.

For my own recollection, plug and socket wiring.
Pay notice: These RS232 (DB9) connectors are not specified for high voltage, so do not take my use as a recommendation!

Pay notice to different filament wiring for ECC83/5751 and 6N2P.
SILK MC trafos (BACK TO TOP)
SILK Supermalloy MC-step up transformer from SAC Thailand: http://www.sacthailand.com/. These are MC220 allowing 1:10 with 100 ohm loading impedance, or 1:20 with 25 ohm loading impedance. Download wiring diagram here. 4-pole switch will be placed on rear panel for 25 and 100 ohms load.
RIAA pics (BACK TO TOP)

Aftermath - and RIAA tweaks (BACK TO TOP)
Being a novice in tube technology, my tube adventures usually create some response from readers more experienced in the art of building amplifiers. This RIAA was no exception. Opinions may be diverse when it comes to loudspeakers, but when it comes to tube amplifiers, people have indeed very strong opinions on what's good and bad. Every designer seems to hold on to certain principles and regard other designers' circuitry with great suspicion, sometimes even contempt based on (anonymous...) statements found on the web. I build what others have done and I listen. Over the years I have built amps based on a wide range of principles and I have borrowed amplifiers to gain experience on sonic signatures from different circuitry. The choice of valves, capacitors and resistors and their impact on the final sound makes an un-endless number of variables. I have faithfully tried a large number of suggestions coming in and I have to say that only a very few paid off in terms of improved sonics. One reader almost choked on his morning cereals when seeing ICEL polyprops for the passive network. Ideally I would have used 1% Philips polystyrene, but my local supplier told me these weren't made anymore, thus the 1% 10n ICEL PPs. I had some Russian 10n silver micas, but wasn't sure they had the necessary precision for this delicate part of the phono stage. Eventually I found 0.3% 10n silver micas from another supplier and changed all 10n caps for these including the grid cap on upper 6C45pi tube. The immediate impression from this tweak is improved sonics in upper mid and treble and despite their bulky size, they stay in. The 2200 uF/10V electrolytic cap on the lower 6C45 cathode was removed and despite lower gain, this did not to any noticeable degree improve the sound, but if less is more, I leave it out. Later, after having changed the ICEL caps with silver micas, I tried inserting cathode caps again, this time 100 uF/16V black gate type N, and realised I missed the loss in gain. Never say never.... Resistors: From what I have been able to pick up from the web, MOX resistors for power (e.g. 1W Beyslag for anode and cathode resitors) and carbon for signal, e.g. grid resistors and resistors in series with the signal - or generally where current is low. I'll see if I can find values suitable for this circuitry. Capacitors: Some people will reach for PiO for coupling caps - I've tried - and to my ears they're not the best. Mundorf silver/golds have been used extensively but were finally replaced by tin foil caps. Whether wound on teflon or polypropylene seems irrelevant from my experience. I have bought a few tin foil caps wound on teflon, but they don't sound any better than their cheaper brothers wound on polyprop. I guess the sheer weight of the tin foil is a major contributor to the sonic signature of these caps. Angela has these up to 1 uF but doesn't ship to Europe. Audyn and Mundorf tin foil caps do the same thing and Mundorf even have 2.2 uF/250V. Audyn e.g. 1.8 uF/250V - and a uF range interesting even for loudspeakers (6.8 uF max). The size of the output cap must be balanced with the input impedance of the line stage and 100k should make -3dB @ 16 Hz if 0.1 uF is used. Some regard this as way to high (-3 dB) and if you have speakers reaching 20 Hz, a higher value is needed. For a 5-10k line stage several microfarads are needed. These 10n silver micas are pretty big and not easy to implement nicely. Reviewing the whole project - and listening to my favourite vinyl recordings - these last months of work seems mission accomplished. I may be building more phono stages in the future to experience other kinds of circuitry, but this phono stage makes a perfect match to my beloved Dynavector XX2 mkII cartridge. Here with silver-gold caps. Not a bad choice either. Measurements (BACK TO TOP) The phono stage was measured with the Hagerman anti-riaa circuit. CLIO output was set to -6.4 dBV to make 500 mV @ 1 kHz via MM input. Left graph: There's a 0.6 dB difference between channels, not enough to make me run for adjustments. I didn't try swapping tubes to find better match, but it may be possible. As can be seen MM input is flatter than the MC input, which is expected. RIAA is basically flat between 100 and 10000 Hz. From 10-20 kHz there's a 2 dB decline. From 100 Hz to 20 Hz we see some 3 dB decline for MM and 5.5 dB for MC. Right graph is displaying the phase of one MC channel (blue line). All in all I'm pleased with the results.

Chassis (BACK TO TOP)
Chassis parts are made from 2 mm aluminum. Making two chassis from 12 pieces of aluminum is time consuming! Left: Main chassis for MC trafos and amplifying board. Right: PSU chassis and yes, the board fits in.
Schematics (BACK TO TOP)