|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
71.50.23.239
In Reply to: RE: here we go again....... posted by vinnie2 on August 19, 2016 at 06:15:35
Hi Vinnie,
My 2 cents. 71A has a mu of 3. KISS rules on ALTECS !!! You can do a Direct Coupled tube amp, a simple two stage amp, if you use a mu of 70 or higher tube. I suggest mu of 100.
I would be able to send you a complete write up, on how to design and build such a circuit, just send me you email address. I've already done the write up.
Have fun with whatever you do !
Low powered tubes like the 71A require one to have great transfer efficiency, ( wiring ), inside the amp, to the speakers, and inside the speakers, to get the most of them. ALTECS are very sensitive to this !!
Jeff
Follow Ups:
Jeff
What I am really would be interested in seeing is your write up on how to design the circuits as this is meant to be more a learning exercise than anything else.
email is vince@vrhea.com
thanks
Edits: 08/19/16
I agree with Jeff: a direct coupled 6K5 to 71 would be the cat's ass. I used a K&K CCS on a 6K5/2A3 amp with the compact cascode CCS set for 1.3mA. Ask Kevin for a 5k ohm pot to replace R2, the 1k. Note the 6K5 will not drive the grid of the 2A3 to 0V as the 2A3 starts to draw grid current about -5V; the 71 may be easier to drive. BTW: on the drawing Channel 2 should be 510V.
May I critique your SET amp's supply ??Power supplies in SET amps ( or P-P ) should avoid any form of a "C" input filter ( 30 uF in your case ).
The ideal tube amp supply really needs to be L1/C1/L2/C2.
Your L1 is too high in DCR by a factor of 2.5 times, needs to be 20 Ohms or less. I like 10 Ohms or less. Also, the inductance is far too high, ruins the dynamics and listening FUN that is possible.
Your L2s are also 5 times too high. 105 Ohms DCR is very poor.
That amp will be DEAD sounding, versus a LSES type supply. !! Compressed dynamics and boring.
Look up LSES posts from eight years ago up here, or, send me your email address as Vinnie did and I will send you my write up on such a amp design and also, ALL the chronological posts by EEs Swenson and hasquin on such a power supply as I suggest.
Going to RMAF this year in Denver, October?? Be sure to stop in at Serious Stereo, where you can listen to ALTEC 604 MLTLs, using a JJ 2A3-40 amp like yours, with, I believe, UNDER 5 Ohm L1s and L2s !!
Have FUN hearing what I suggested, ..... I am !!
Jeff Medwin
---------------------------------------------------------
ADDENDUM,To make it easy, I have included a simple discussion of LSES principles by E.E., and AA Forum Member John L. Hasquin, that I quite like:
A choke's behavior is to oppose change in current flow. A cap's behavior is to gobble up and store as much current as it can. When the amplifier demands an increase in current the caps initially release some current, but then begin to compete with the amplifier for current because they want to maintain their charge.
Then if you have large inductance chokes, they prevent the power transformer from delivering current to the caps because they are opposing change in current flow. This in turn means the more current the
amplifier demands, the more the caps compete for the limited amount of current.The result is dampened dynamics because the amplifier is current starved. Big caps and chokes = dampened dynamics and a current starved amplifier. Low DCR power transformer with low value chokes and caps = dynamics.
Edits: 08/20/16 08/20/16 08/20/16
Hi Jeff, yes I know that the supply is not ideal, but I have been using this same chassis for years as a test mule and the 105 ohm transformers fit in the available space. I have plans build another set of monoblocks with low DCR transformers and even have the parts, and have drawn the top plate CAD drawing for Front Panel Express, but for now I am experimenting with the old war horse. I wouldn't be using it if it didn't sound good, but I do have plans for a new set of monoblocks.
BTW: you sent me that pack of LSES posts YEARS ago.
In October I will be in Amsterdam listening to live music at the Concertgebouw.
.
Let's not make it more than it is Jeff. I am just curious to see what procedure you use to come up with the major design points of a two stage SET amp.
While I'm not against CCS's as plate loads in general, I can't see the benefit in this application. MOSFET's at such low currents exhibit very low Gm and even though putting them in cascode multiplies their Gm, it's still too low for all practical purposes. BJT's would offer better performance here I think and a gyrator even better. In any case, why not just use a plain ole resistor as the load? Don't get me wrong, I'm not saying that this circuit doesn't sound very good (I'm sure it does), only that it might also sound very good with resistive loading while reducing complexity.
However, the biggest issue here IMO, is that the proposed amp using the 71A as the final, will be very low in power. The output tube will be running at a plate dissipation of .02x180=3.6W and If we assume a plate efficiency of about 20% it will have an output power of 0.7 watt. What's the sensitivity of the OP's speakers?
I agree that there are many ways to get the sound one is searching for. While the CCS I chose may not be ideal, the proof is in the pudding. I tried several alternatives and this approach was marginally cleaner and clearer than the alternative approaches. Maybe it is the 6K5, maybe the CCS, I can't really say other than it is slightly and noticeably cleaner than the other topologies I tried. Why would a 6K5, a low transconductance tube from the 1930s, sound cleaner than a state of the tube art high transconductance Mullard E180F (triode connected) from the 1960s?
As an aside: I use the K&K CCS because I tried wiring up MOSFETs on perf board but was frustrated with the perf board, small parts and solder "wiskers" and I really couldn't see well enough to ensure I was soldering what I thought I was soldering. The K&K CCS board and Pete Millet's A2 buffer make using solid state devices easy. I know I can experiment with different devices like the BJTs on these boards, but for now I am using them as designed because they sound excellent. I have plenty of alternative SS devices for my next amp using the 6F5 and will experiment some more.
As always the proof is in the pudding as the saying goes and results trump theory every time. I also don't like to fuss with perf boards and small parts and so I understand the appeal of ready-made SS boards. This convenience along with your stated performance gains, I can see the appeal. Perhaps I'll try it myself in the future.
They are Altec Granadas and are about 91 db I think, maybe a little better. I have run them satisfactorily with quite a few flea power amps. Not window shakers, but then I don't like loud music anyway, just good music.
I'll agree and say that I would take quality over quantity 99% of the time. However, in this situation you might be pushing it a little bit with a low 90's speaker and less than 1 watt of available power. Not saying that it's not going to work at all, but you will be limited IMO.
As I told Jeff, I really want to learn from this. The post you did on the quick and dirty way to determine operating points was one of the things I used to get these for the 71A. Got anything like that handy for picking driver tubes? : )
First you need to determine the operating points for the output tube. That tells you what your plate supply requirements are and also what the drive requirements are for the driver. A 71 operating at max plate voltage of 180V will require 41V to drive the tube to clipping. A typical source has a 2V output so no preamp should be required; if plan for an input voltage of 1V to drive the amp to clipping, you need a driver with a gain greater than 40.
Some tube manuals have a chart showing tubes grouped together by their amplification factor. I chose the 6K5 because there seems to be something special about these high mu tubes with their cylindrical plate and their concentric cathode, grid and plate. My next amp will be 6F5 driving a EML 320B using K&K CCS.
How did you determine that 41V will be required to drive it to clipping? Those are the kind of things I am hoping to learn.
"How did you determine that 41V will be required to drive it to clipping?"
I learned it from my tube manual which has the operating conditions. Notice the operating points Ep = 180V; Ip = 20mA; Eg = -40.5V.
Since the max signal we can input the tube is 40.5V peak; 40.5V peak or 28.6 Vrms should drive the tube to clipping. That's not much of a demand. There are many tubes that can accomplish this with 1V input. If you choose a lower plate voltage the grid bias will be less and the max input voltage will be lower. As has been pointed out above, there is not much power available with the 71A. I had thought to make a tweeter amp with mine, but went a different route.
Thanks for the explanation. I actually had a 71A amp hooked up to these speakers sometime ago and it was adequate for my listening needs. It is also a very nice sounding tube. I want to try some different drivers to see what they sound like and to try and learn some design in the process.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: