diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Tubes / Valves (http://www.diyaudio.com/forums/tubes-valves/)
-   -   Power tubes in cascode for output section? (http://www.diyaudio.com/forums/tubes-valves/213608-power-tubes-cascode-output-section.html)

Austin Translation 31st May 2012 02:57 AM

Power tubes in cascode for output section?
 
Has anybody ever tried this to see what it sounds like? I think it's pretty interesting, especially as a super easy way to drive the top tube into grid current.


Power cascode - Power Cascode

Austin Translation 1st June 2012 04:36 AM

Ok heres one idea, since I already have a single ended el84 amp pulled from an old columbia console that I experiment on- I pull the tube rectifier and use 1n4007s instead.. Then use the 8 pin rectifier socket for a dual triode to drive the el84s. Im thinking maybe a 6BL7? Obviously a twin triode that had a 5 volt heater would be best but I don't know of any. Maybe I could just try it mono first to see how it sounds with one el84 on bottom and one on top so I don't have to wait for tubes in the mail to find out if it will even work.

SY 1st June 2012 08:06 AM

There was an article (two part) by Chater in AudioXpress a few years back showing a few amazingly complex cascode output circuits. Worth reading (sorry, don't have the exact reference at hand, but you should be able to dig it up).

A major issue with doing cascode output stages is the very high source impedance. Transformers don't like that- much feedback required..

Kay Pirinha 1st June 2012 09:30 AM

The very first Music Man guitar amps were constructed this way, i.e. a transistor (MJE fifteenthousandsomething) in the cathode lead of each 6CA7 power tube.

Best regards!

Austin Translation 26th June 2012 07:48 AM

Just a follow up, I modded my magnavox like this using solid state diodes instead of the 5y3gt and in its place added a 6080 to drive the el84s. It works well but I want to go with fixed bias on the 6080. I use a 4 amp wall wart to power the el84s and 12ax7, and the original heater output for the 6080. http://farm8.staticflickr.com/7116/7...193d1d6f87.jpg
crazy amp2 by Dirtyrottenstinkingbiker, on Flickr

Austin Translation 26th June 2012 08:42 AM

http://farm9.staticflickr.com/8026/7...87b69323_b.jpg
crazy amp2 by Dirtyrottenstinkingbiker, on Flickr

DF96 26th June 2012 08:52 AM

I can't quite see the point of this. Why choose a circuit which guarantees low efficiency and high output impedance, when you usually want the opposite in an output stage?

Austin Translation 26th June 2012 02:09 PM

Quote:

Originally Posted by DF96 (Post 3072870)
I can't quite see the point of this. Why choose a circuit which guarantees low efficiency and high output impedance, when you usually want the opposite in an output stage?


Could you please explain why the output impedance is higher than a normal se el84 output stage?

To my understanding, the output impedance here is mostly dependant on the bottom tube. Isn't that correct?

One point of it is to be able to run the el84 in class a2, which should increase efficiency right? Another reason I did it is to use this as a learning tool, to see proof that it will work.
I am getting the same 6 or 7 volts ac across an 8 ohm load as I was when it was simply pentode el84, but it is trioded now. Before with the el84 as a triode I was lucky to get 4 volts. I believe it may get even more power if I can make a decent screen supply, especially if I can get the bias just right. Either with a cathode bypass cap and maybe a different value cathode resitor or even fixed bias. Thanks for the reply!

DF96 26th June 2012 03:42 PM

The output impedance of a cascode is roughly mu(top) x Ra(bottom). This may or may not be higher than a pentode, but it will be much higher than a triode on its own. An unbypassed cathode resistor will add mu(bottom) x mu(top) x Rk.

It will get you into A2 mode, but at the expense of a high cathode voltage which limits efficiency. Interesting as a learning tool, but in most cases I would think that more conventional outputs are better.

The main benefits of a cascode are low input capacitance and pentode-like gain without partition noise. Both are very useful for video and RF, but less so for audio.

the_manta 26th June 2012 04:24 PM

This might by useful: Improving the Cascode's PSRR (page 2)

As DF96 said, this yields a very high output resistance. Cascode is nice for input stages (be careful with upper heater supply and bad PSRR) but you won't get a decent output stage with it. There are better "tube above tube" concepts for that like White Cathode Followers for example.


All times are GMT. The time now is 08:02 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2