|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
72.235.7.37
Is it too much to ask of the lm317 cvs to handle the cathode biasing duties of a pair of 7591 in push pull? I like how they perform in my SE amps and wonder if they can work in an EICO ST40. Thanks for your suggestions. cheers, Dak
Follow Ups:
If you are planning to operate the amplifier Class A, current source biasing the output stage is a fine idea. If on the other hand you are staying with Class AB (which I'm assuming is how Eico intended it) current source cathode biasing won't work. In fact cathode self bias in a Class AB stage is a compromise even with a cathode resistor.
Phil
I think a regulated voltage at the grid or cathode might be preferable to using the LM317 as a current regulator. With a device like the LM337 (as a cathode voltage regulator), the amplifier can be operated in any class the designer chooses. In addition, variations from one tube to the next and changes due to aging will probably produce fewer issues. The question is, how to keep either of these solid state devices from influencing the sound. A better choice IMO is to spend a few dollars more and add a small negative bias supply. This eliminates the problem of ICs in contact with the signal path, does away with need for a (usually electrolytic) bypass cap, and produces maximum power (headroom) from the tubes.
--------------------------
Buy Chinese. Bury freedom.
Edits: 02/20/15
http://tronola.com/A_New_Look_At_An_Old_Friend.pdf
Maybe LM 337 would suit even better.
ecc230
Is this the enhanced bias from DCG? Very nice, but again added complexity and the need for adjustments.
The mod is for a unit that I am prepping for a friend who is not comfortable with going "under the hood" as it were.
Maybe, it would be best to just leave the unit stock. cheers, Dak
The LM317 generates a significant level of broadband "white" noise. In the April 2009 AudioXpress letters section Walt Jung Reports that the LM317 at 10mA had 50 times more noise than a DN2540. Since the common cathode node of a PP amp is not an input that might not be too much of a problem.
How much bias current will pass though the current source? If it is less than 90mA you could use a pair of cascoded MOSFETs. W. Jung writes in the above reference: "...the (cascoded MOSFET) configurations are the most versatile of any discussed. I recommend them without any major reservations." K&K Audio sells a nice kit for little cost. My phono stage uses 17 of these, 4 are used in the line stage and four in each mono-block. They work very well.
My mono-blocks have 300ma of output stage bias. I use a TL431 based current sink. The circuit is shown in the Fairchild data sheet. In place of the transistor I used a power N-channel MOSFET. Instead of the bias resistor connected to the cathode I used a LM334 /DN2530 cascode set to 2mA to bias the TL431. Fairchild makes a low tempco version the KA431. My bias comes up on startup and sits at 300mA and doesn't budge. Cold or hot.
"It is better to remain silent and thought a fool, then speak and remove all doubt." A. Lincoln
It's an easy task, but depending on the goal could lead to different results.
You can use one LM317 to bias both tubes in a PP arrangement, and you can further decide whether to bypass the LM317 or not. This will set max current for the pair, but the tubes still need to be pretty well matched for optimal results.
You can otherwise use one LM317 for each tube in a pair, achieving perfect DC current matching from un-matched tubes, but you have to bypass each LM317 and are going to be limited to class A operation (neither device is going to be able to switch off because of the constant current draw.
If you have experienced improvements in SE amps, the second approach in particular is going to yield important improvements in PP operation, at the "cost" of limitation to class A operation.
******
http://rh-amps.blogspot.com/
Hi Alex. Limiting the tube to class A would that mean any other changes ? Or it would just mean that adding one LM317 to each output tube + bypass cap? cheers, Dak
Hi DAK,
No other changes than those explained above in my post. Basically, the main difference in operating a PP amp is whether any of the output devices can switch off during some part of the operating cycle (class AB or B) or the output devices never switch off during any part of the operating cycle.
If you use one LM317 for both output devices (tubes) the pair will draw a set constant current, but either tube will draw as much current as it "wants to" statically, and dynamically one of the devices can switch off - therefore possible class AB or B operation.
If each device has its own LM317 (as if it were 2 SE amp output stages) each device will draw a set constant current and will never switch off: therefore class A operation.
Logically, class A operation yields less output power than would be possible with class AB or B, and without entering further into details like what would be the optimal primary impedance, current, etc. for a given output stage in a given class... it is more than obvious that if you plan on modifying some amp by inserting a CCS device under the cathode(s) of output tubes, this is most probably going to affect output power as well.
I am gladly going to pass on commenting the LM317 as suitable or not for CCS duty under the cathode of output tubes, and the white noise thereof, etc. I gladly leave this type of discussion to people I like to call "armchair DIYers" (no pun intended) who build one (new) amp for every approx. 10k posts they make. That's a good way to keep oneself busy - posting without building, and it helps getting things into the right perspective (like, white noise and other stuff people who build real life amps relatively often actually do not tend to notice all that much)...
******
http://rh-amps.blogspot.com/
Hi DAK,
If you want to go Class A you will lose some output power in order to stay within the dissipation rating of the tubes. You do want the cathodes connected together and NOT bypassed with a capacitor. This way the even order non-linearity of the tubes will cancel.
Stay within the ratings of the tubes and the current source which you use.
You may get some improvement just by replacing the cathode resistor with the current source. If we know the turns ratio of the transformer and your speaker impedance, and can lower the plate voltage, the operating point can be optimized to give you the most power you can have staying within the rating of the tubes.
Phil
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: