In Reply to: LCR - RIAA posted by Gnnett on June 7, 2003 at 15:30:30:
Koinchiwa,I'm not sure what Brian does, but a few remarks. S&B is currently well under with OEM jobs plus the making of current orders. They have more work than they really can handle, hence development is slow.
The 600R RIAA's exist and are pretty readily available (and measure absolutely peachy - my first rejected set had 20Hz-20KHz +/-3db depending upon level - the latest are +/-0.1db over the same range), the 10K are not yet "canned" (in either sense of the word), but there is just no development time, so the 10k will be some time in the comming if at all. And considering that a 10k RIAA will be at least 16 times as hard to get right than a 600R RIAA it will likely take a lot more development than was spend on the 600R RIAA.
Now as to making a Phonostage with the 600R RIAA or modifying existing units....
You can either design from scratch. To drive the 600R RIAA you can choose a few options. I have discussed before in various places, but as Brian sells the 600R RIAA and as the information posted on the S&B pages is perhaps a little too abstractly technical for people I might as well put a few comments here (Brian, feel free to edit and repost at the DHS Site).
The first thing to understand about the 600R LCR RIAA is that it behaves in essence as frequency independent 600 Ohm Resistance load to the source, IF the RIAA is terminated into 600R (which is the normal use).
So, ANYTHING that is happy driving a 600 Ohm resistive load is happy to drive the 600R RIAA. Of course, virtually NO Valve circuit is happy with such a load.
One option is to use some from of cathode follower output and a large value (50uF++) output coupling capacitor to drive the RIAA. I have seen schemes using White Followers, normal Followers, modified SRPP, you name it. Of all those I liked best the use of a simple Mu-Follower....
A popular and very workable approach is to use a 6DJ8/6922 Mu-Follower as seen below within an rather interesting Phonostage from Japan (by Jun Sato):
![]()
The Choke in the Mu-Stage is "gilding the Lilly" and if desired can be replaced by a resistor, though it's a truely neat touch if you ask me.
In fact, the insanely expensive Wavac LCR Phonostage seen below is no more than an external solid state regulated powersupply, a pair of decent MC Stepups followed by a 6DJ8 Mu Stage, followed by the 600R RIAA and followed by another 6DJ8 Mu stage.... The Phonostage can be seen here:
A Circuit that will be damned close to the Wavac, if it is not identical (as both it and the Wavac Phono where designed by the late Nobu Shishido) is here:
Now take this circuit, preceed it with your favourite MC Stepup Transformers (I recommend the S&B TX-103, items from other manufacturers are available), fit the LCR RIAA and you likely have a phonostage that is better than maybe 90% of what's out there, especially if a good quality PSU is used. I personally would be tempted to go for a free-running, non-looped Switched Mode supply, but that's just me....
Used without MC Stepup the gain of the Phono is around 38db, add a TX-103 for another up to 26db gain and you have 64db gain for the Phonostage. The output will happily drive the S&B TX-102 passive transformer attenuator as used in the Django and it will do so even with 6db stepup, so from the TX-103 in to the TX-102 out you could have a total gain of 70db.
This is about the same as the CAT SL-1 Preamp, so any system where a CAT Pre works could be changed over to the above Phono with TX-103 and TX-102 passive line with 6db gain. Do I think the resulting preamp would better the CAT for a still fairly modest outlay? No comment... ;-) !!!
For the adventerous, you could build also a hybrid Mu-Stage as shown here (includes here also a Stepup Transformer already wired in - though this circuit is meant as Mike Pre):
This circuit is from the Welborne Labs "Hybrid Mu-Stage" Kit and can be used with either ECC83/12AX7 or ECC88/6DJ8 Valves. I suppose the recently much talked about IXYS (if I got it right) depletion HV Mosfets K&K is using in active loads could be used taking the output from the Source of the Mosfet. Anyway, plenty of options here but you need again a whopping big coupling cap to the 600R RIAA.
The other often taken route is to use a 4:1 or 5:1 stepdown Transformer (often in Parafeed) and use it with a 417A/EC8010 "Class" Triode. The loss in gain due to the Impedance matching transformer means that you usually need another 1:4 Transformer after the RIAA to get the gain back.
If you used the russian 6S45-PE and a S&B TX-103 inverted as 5:1 Stepdown you have after the RIAA in effect something close to unity gain. Following the RIAA with another TX-103 in 1:5 stepup gets us 14db gain back, so if then use another 6S45PE with another TX-103 as output transformer we have a Phonostage with a gain of around 34db gain and a very low output Impedance of around 50 Ohm, but at the cost to have with the MC Stepup a total of 4pcs TX-103 per channel.
Sure it can be done and you may wish to procure also 4 pcs of the rare but occasionally sighted cornell-dubilier 2u2 Silver Mica Capacitors to use as Parafeed Coupling Capacitors to the Stepdown TX-103's and while we are at it, order the TX-103's in silver and get a silver special RIAA. Bring a Platinum Amex card and your chequebook though.
I'll not illiustrate this approach here for completeness with a schematic from Joseph Aigner drawn professionally using the "Wobblyhand (TM)" schamtic editor, but I find this just a little to "Heavy Metal" for my taste:
You can use slight variations of the above principle, such as a 2-Valve Frontend using 6072A and 5687 in a DC coupled Amplifier stage followed by a 4:1 stepdown transformer, then RIAA, then another 5687 again with 4:1 output, giving overall around 36 - 38db gain with one less transformer and one more valve stage and many more.
Now the next trick is, I will readily admit equal to cheating, but so what. It has the advantage that it can be used to convert ANY passive EQ Phonostage, no matter what valves are used and how it is arranged etc to using the 600R EQ. The only price for that - your soul (just kidding). Or more precisely the use of Solid State Buffers (to many this IS selling your soul).
If you simply take a pair of BUF03 (I know discontinued) or discrete J-Fet Buffers as per Borbelys Article and feed them +/-12...24V (depending what they are rated for) and feed them an already DC blocked signal (a 10nF ACL Silver Mica Coupling Cap and a 3M3 "gridleak" resistor would be my choice) you can get an output that allows the 600R RIAA to be coupled directly and to be driven from a low impedance (< 100R).
Borbelys Article is here:
http://www.borbelyaudio.com/ae699bor.pdf
Look for figure 15C or 16A (or also 16B but that is harder to get right) and use the 2SK389 Dual J-Fet. Often you will be able to use +/- 12V Supplies (can often be generated from the heater windings) and omit the resistors in Figure 15C if you use the arguably expensive (but compared to other options fairly cheap) Monolithic Dual FET which should show a zero output offset for the follower.
Of course, such approaches to using an affordable Solid State buffer can be applied to both existing design and new builds.
I hope the above notes help those who want to "try" the 600R Constant Impedance LCR RIAA. I would add that I have considered all the schemes mentioned above and a few more, breadboarded some and in the end have used non of the above, but that's just me. My own Phono has been extensively discussed in my Technical Yahoo group, so I save you the agony of having to read about it again.
Sayonara
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- Re: LCR - RIAA - Kuei Yang Wang 12:28:09 06/11/03 (2)
- XLNT Info GR8 HLP Thx (nt) - Gnnett 21:25:48 06/13/03 (0)
- Thanks KYW - This ones a keeper! (nt) - bcherry 16:16:09 06/12/03 (0)