| gema |
Has anybody tried something like this? Any referencis? Anything?
Regards. |
|
|
| GRollins |
There's certainly no reason not to use a tube preamp for a Zen...I'd say go for it.
If you're building them as a dedicated pair, you could get rid of one coupling cap.
Grey |
|
|
| mg16 |
If you mean the original Nelson Pass Zen amp, you may have a problem with a tube preamp. As I recall in his article, the input impedence is very low, and tough to drive. That is partly why he followed up with a suitable ss preamp for it in Glass Audio. Zens input Z is Something like 4.7 kohm.
A tube preamp can swing a lot of output voltage, but current output is required for an input impedence this low. Preamp tubes can't source much current at all, and there output impedence is typically high.
2 Potential problems if used with a low z input power amps.
MG16 |
|
|
| GRollins |
I don't know as I'd use a 12AX7, but a 6DJ8/6922 can do this easily. There are other tubes that would probably do just as well.
Grey |
|
|
| alaskanaudio |
I used to drive such low Z loads with the old totem pole output follower stage which is now likely now known by various other names. As I recall I used either both sides of the 12AU7 or the 12BH7 depending on the application. Driving loads down to around 600 ohms was possible with a suitably sized output capacitor.
This of course was many years ago and I don't remember the exact particulars. But If I were the build something like this again it would likely be my choice of circuits.
The difficulty of these or similar circuits is that the cathode of the upper section floats well above ground. Thus cathode to heater voltage ratings have to be watched or shorts can occur between the two. One way to attack or reduce this problem is to apply a DC bias to the filament supply and float the fillament supply at level that is some where between the upper and lower section cathode voltages. The other is to use a seperate filament supply for the upper section.
John Fassotte
Alaskan Audio |
|
|
| DrewP |
Funny you should ask, I'm currently in the middle of work with a friend building a choke loaded single stage no neg FB SE preamp using a single EL84.
In theory it ought to work fine, give enough voltage gain and feed the evil 4K input load with no hassles.
As i say, work in progress but I"ll post when I have results |
|
|
| gema |
:D
Very hapy to hear that! I am realy looking forward to hear more from you. Good luck with the project!
Regards |
|
|
| lilolee |
| quote: | | I'm currently in the middle of work with a friend building a choke loaded single stage no neg FB SE preamp using a single EL84. |
Sounds really interesting. That should be able to drive most things. Can't wait to hear more.
Lee |
|
|
| Nelson Pass |
Keep in mind that the Zen circuit itself has near
0 input impedance, being a virtual ground. We get
the impedance simply by inserting the input
resistor in series.
The good news is that it can be driven by any source
impedance, in fact, it performs better driven by a
1 ma or so current source. This means that any tube
output circuit that can come up with 1 ma off the plate
will drive it just fine, regardless of the plate resistance. |
|
|
| DrewP |
As Nelson has said, if we can feed upwards of 1mA we'll be fine. We're working on a 20mA bias point at present with a 5k attenuator post the coupling cap. That side of things ought to go well.
The concern (and we're being paranoid here) is that the EL84 mightn't be the best sounding tube for the stage. Having read that Tim d'ParaVicini's V20 amp using parallelled preamp tubes as output tubes sounded better than using the usual suspects
(EL34, EL84, 2A3, 300B, KT66 blah, blah, blah,) means that maybe we're going in the wrong direction. (power used for pre that is)
Parallel pre tube single stage would just give us a Conrad Johnson Art clone I supppose.
I guess it's in the application.
Initially most of the components (specifically the resistors and socketry) will be medium level carbon films and so forth. I'll put in some good wire, coupling caps and cathode bypasses but I won't be doing version 1 with vishays, Jensen copper foils and goldpoint attenuators. We'll see how it sounds before we tweak using stratospheric bits.
Time will tell, some construction work will have to wait till this months (and next months) bills are paid.
Waiting is...
Drew |
|
|
| lilolee |
I've been try to track down a schematic for a paralleled pre amp tube power amp, but can't seem to find one. I have about a dozen 12AX7 lying around so wouldn't mind giving that a go, or are they too high a gain?
lilolee |
|
|
| DrewP |
Originaly as a tribute to English Rolls Royce and Jaguar engines, Tim dPV was planning to produce the amp as V12 not V20. Sept 1998 issue of hifi news and record review says that the Yoshino V20 uses 20 ECC83/12AX7 ie 10 parallel sections per phase.
Haven't seen a schematic anywhere. Aside from the power output, I can see no reason not to use 12 tubes and 6 parallel sections per phase.
Not sure what the implications are for output impedance.
Beyond my design abilities I'm sorry to say but it sounds like exactly what Tim originally planned to do. (in fact the article indicates that the prototype was configured precisely this way with 12 tubes not 20)
Design is class A push pull.
Can't help much more than that sorry.
Good luck
(more point to point wiring than I like, fear I'd produce a rats nest!)
Drew |
|
|
| GRollins |
Tubes parallel very nicely.
You might want to put a grid stopper resistor in series with each grid, say, 100 ohms to 1k.
You have two fairly obvious choices:
1) Parallel all the stages together. The advantage here is that the Zout will be lower.
2) Run each tube as a differential, then parallel all the differential sections. The disadvantage is that you won't have quite as low a Zout. But you get several things in return. Differentials are inherently lower distortion. They exhibit good CMRR. If you use a current source to run them, you can convert unbalanced signals to balanced. Or, conversely, you can take the output from the 'backside' of the differential and it will be phase correct.
As you might suspect, I vote for option #2. Once I get around to rebuilding my line stage (Ha! Probably sometime next century at the earliest.) I'll be doing something along these lines...with a few embellishments, here and there.
The 12AX7 is a high mu tube--100, if I recall--which gives you a theoretical maximum of 40dB gain. In the real world, and assuming that you don't start cascoding and such, you'll end up somewhere in the low 30s. That's more than most folks will need, unless they're running a Zen, or SOZ, or one of those 'follower' amps, which could use the help. Options abound:
--Use a voltage divider at the input to reduce the incoming signal.
--Use a pot in place of the voltage divider.
--Use a few dB of feedback to touch up the performance of the circuit.
--Use a fixed resistor to pad the volume control down a few dB.
Jeez, I dunno. Lots more possibilities. Mix and match. Use your imagination. That's why Horowitz and Hill called it The <i>Art</i> of Electronics. Just get out your palette, your brushes, and your jaunty little beret and call yourself an artist...
Grey |
|
|
| GRollins |
As a preamp, or as an amp? If you're planning on using it as a preamp, it's overkill...the Zen isn't <i>that</i> difficult of a load to drive. But I see no reason why is shouldn't work well. Sometimes a little overkill is good for the soul. Which reminds me--I had this idea the other day for an amp circuit that...but lemme get the Aleph-X done first.
By the way, I've got a schematic drawn up for a tube X preamp if anyone is interested.
Grey |
|
|
| Peter Daniel |
Grey,
I would be interested. Search for a perfect preamp ( if there is such thing) never ends. |
|
|
| gema |
| Count me in Grey. Thanks. |
|
|
| psarin |
just interested if anyone has tried this combination? i read the above comments about 12AX7's maybe not being up to the task i am looking at a really low watt application (2-4W monoblock SOZ's with a PAS-2?).
thanks in advance for any tips, impedance matching advice! |
|
|
| GRollins |
Honestly, the Zen and SOZ aren't that difficult to drive. When all else fails, try it. It'll probably work.
Grey |
|
|
| HarryHaller |
I don't think so!
1. Tube preamps usually have 0.5uF to 2uF output caps. Pretty high low frequency corner into 4.7K ohms (Zen) and 1K ohms(Zen revised.
2. Tube preamps often have output impedance in the 500 to 2000 ohm range. You are throwing a lot of gain away into 1000 input impedance and making the output stage of the preamp work even harder.
3. Output stage biased in the 0.5 to 3mA range for tube preamp. For low distortion you want your signal current to be a small fraction of the bias current.
4. Son of Zen has pretty capacitive input. Limited H.F. bandwidth
and increased H.F. distortion with tube driver. See item 2.
5. Probably work? Probably not real well.
Pass the Grey Poupon, I don't think tube preamps are going to cut the mustard with Zen amps. Should work better with Aleph type circuits, better if you change the input impedance to 100K for
the non inverting input of the Aleph amp.
H.H. |
|
|
| dshortt9 |
I would propose a Constant Current Draw stage as shown at the bottom of this link:
http://www.homestead.com/whaan/files/index.html
This circuit can be used to good advantage with the 6H30 which has a heater to cathode max voltage of 400. This eliminates the need for a second floating filament supply. |
|
|
| Nelson Pass |
Certainly there is some challenge for tube circuits
to drive the Zen, but here are some items to consider:
1) The output impedance of the tube preamp is not
much of an issue since it ony reduces the gain, and in
fact will give lower distortion, all other things being equal.
2) The output capacitance of a tube preamp will induce
a low frequency roll off based on the output resistance plus
the input resistance. Most of the time we see 10 uF,
and with a 1000 ohm output and a 1000 input, the corner
frequency will be below 10 Hz. Not great, but do-able
in many cases, and you can always put in a bigger
cap.
3) If you want higher input impedance, just use higher
values for R10 and R11 (revised circuit). With 4.7K input,
we see some more distortion at 20KHz, but no worse than
the original Zen.
4) Output current of the tube circuit is the main thing, since
they usually have plenty of voltage gain to throw away.
To drive the Zen to full output, we are talking a tube output
stage which can cleanly deliver 1 mA for the original Zen
and 4 mA for the revised Zen. Cathode followers are called
for here, not outputs directly off the plate, although that, too,
is do-able.
5) Upcoming Zen project will feature a much higher input
impedance, and should be adaptable to previous versions. |
|
|
| dshortt9 |
The link that I mentioned previously takes you to an index. The title you want is:
Simulation driver stage capacitor coupled hybrid amplifier using TubeCad®. |
|
|
| gnomus |
| I had no trouble driving a Son of Zen at 50 watts or less using the single ended connectors of a Conrad Johnson PV9A pre amp that is over ten years old. Even the built in phono section driving a low .4 millivolt moving coil could do the job. And the phono uses 12AX7 tubes. What happens is the volume control gets turned up near the 12 o'clock position or slightly more. With your ear near the tweater you hear more tube hiss. Other wise nothing strange happens. It still has that classic full tube sound of older tube equipment. Off course I have changed the tubes once in ten years. |
|
|
|
