Welcome! Need support, you got it. Or share you ideas and experiences.
71.236.58.161
In Reply to: RE: One more point on balancing DC posted by artsybrute on May 05, 2009 at 13:21:53
don't you have to concentrate on the details of a phono pre if you want it to work? ;-)
The Blumlein Garter is pretty straightforward, at least the version that uses only resistors is. You are using individual cathode bias for each tube. Figure out Rk and split the value in half. Tie these two in series from cathode to ground. Now tie a 100K or so resistor from the grid of V1 to between the two cathode resistors for V2. Do the same for the grid of V2. Done.
You can easily add an ammeter to each cathode to monitor (or not). If you do this plus the caps, you can check to make sure the idle current has been balanced by the garter then simply remove the caps and you are done. You could even add/retain the balance pot, but that's an unnecessary expense IMO.
It will work however it gets done. It just won't work as well :^]
According to the values shown in the link, I still have to adjust for tiny currents. So I would still need to put in a balance pot. After that I want to use a single current sink for the push pull pair. And there's a ton of new stuff in this rev, so I just want to get it up and running first without getting fancy.
Maybe it's not so bad. If I can figure out the theory a little better by the time I start wiring that stage (the sockets are already in), and I figure out where to put the pot and how to adjust it, then I'll do it. But first I need to make sure that I won't get grid current if something goes wrong in the cathode web. There are some mighty fine wires in the SUT. I vacillate. No I don't. Yes I do.
Just don't tell me I have to build a whole microphone next.
Triamp... Take a load off!
you're trying to get well below 100 μA. If your tolerances have to be that tight, I would either get a transformer spec'd for this use or go back to a SE circuit and use PJ's points on getting the noise down with cables and step-up location. I'm not comfortable that any scheme will keep things consistently in the single- or double-digit μA range. Do you relly want to constantly fretting over this?
(BTW, I need to do a little more back-of-the-envelope work, but I think that the sum of the cathode resistors used in the garter circuit should be about half that calculated for normal cathode bias. The remaining grid voltage is supplied by the garter, which is where the balancing comes from. With precision NIWW resistors, you should be able to get the tubes very close. Just don't know if you can hope to get them <100 μA close.)
Do you really want to constantly fretting over this?
Um, no. Thus my answer in the post:
http://www.audioasylum.com/forums/bottlehead/messages/13/135758.html
get a transformer spec'd for this
Well, that's the Sowter 9055. Not expensive, but times are tight, the 9055 specs are not as good as the 9063, and the I have the 9063 installed already. That's why I figure I'll just start with the 9063's and caps, perhaps tweak later.
Just don't know if you can hope to get them <100 μA close
Yup, that's the question, if one adds the qualifier "without making it the focus of the whole project". This started out as a simple PS upgrade, went to SUT's, new OT's, direct coupling, then balanced input, now PP first stage, then optimization for headphone output, then a stereo/mono switch at the output along with the headphone/crossover selector switch. Along the way that led to tonearm rewiring, a SAMA, and now a new motor pulley for a flat tape belt like yours. Eliminating the coupling cap interests me, but not if it becomes a new tangential challenge. I wanna hear some music!
But I'm really glad you're working on it. I'll probably contact you about it in the fall, when I'm ready for some electro-nanocurrential-techno-logical adventurism. The real interest to me is the whole concept of using ingenuity to allow for the installation of less expensive, great sounding trannies spec'd for no DC, in place of larger DC rated trannies that have previously been the only choice in these applications.
Thanks again.
Triamp... Take a load off!
over on Tube DIY. (Aside: when did DIY equate to playing with capacitors?) Looking at the garter circuit, I can't see any way that there's any DC on that primary.
It looks like getting the Rks spot on will require a bit of experimenting. You are going to get some bias voltage (maybe 40% to 50% not accounting for insertion losses) from the extra B+ voltage the garter draws. I'm thinking you want to match the resistors exactly and not get into achieving values through additional series or parallel networks, since the resistor match is critical. The extra B+ through the garters makes up the rest with any residual shunted to ground. For example, biasing a 45 at Va = 250V, Ia = 30mA, Vg = -50V gives a Rk of 1k67. I would think two 500r0 resistors in series with the 100k garter tied between them *should* get the current draw about right and balance the tubes. Raymond Koonce has told me that he got the bias correct when he used two 600r0 resistors for the 71A. He's running at the data sheet OPs of ~180V, 20mA, -40V which would require a Rk of 2k0 in standard cathode bias. But apparently 1k2 works.
More cypherin' to do. Later.
Post a Followup:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2009, Audio Asylum, all rights reserved ]
[ Go Green ]
[ Take5 ]