Tube pedal guys - so what about Cascode for a very high overall level of gain?

Started by frequencycentral, July 29, 2009, 07:51:53 PM

Previous topic - Next topic

frequencycentral

Just been browsing Valve Wizard (again!) and it occured to me that we haven't tried anything like this yet with starved plates or with charge pumps / SMPS's in a pedal design yet.

http://www.freewebs.com/valvewizard/cascode.html

"The cascode is a means of using a pair of triodes to achieve very high levels of gain and input sensitivity from a single stage.......the lower triode amplifies in the normal way. The amplified (and inverted) signal is passed from the lower anode to the cathode of the upper valve.......the upper triode has a normal resistive anode load, and so has fairly normal level of gain and does most of the work. In this way, the gain of the lower triode is multiplied by the gain of the upper stage, resulting in a very high overall level of gain!"

Hmmm, I wonder.........anyone tried anything like this in a pedal design? Time to warm up some tubes me thinks!
http://www.frequencycentral.co.uk/

Questo è il fiore del partigiano morto per la libertà!

rotylee


frequencycentral

Yeah, that's what I was thinking. I wonder how a cascode would perform at low voltages.
http://www.frequencycentral.co.uk/

Questo è il fiore del partigiano morto per la libertà!

Ripthorn

Rick, I expect you will also try a dual cascode setup (cascaded cascodes).  That would mean uber gain.  Let us know how it works, seeing as how time to actual sit at the  bench is almost non-existent right now.
Exact science is not an exact science - Nikola Tesla in The Prestige
https://scientificguitarist.wixsite.com/home

km-r

Quote from: Ripthorn on July 29, 2009, 09:08:13 PM
Rick, I expect you will also try a dual cascode setup (cascaded cascodes).  That would mean uber gain.  Let us know how it works, seeing as how time to actual sit at the  bench is almost non-existent right now.

TUBE BSIAB2?
Look at it this way- everyone rags on air guitar here because everyone can play guitar.  If we were on a lawn mower forum, air guitar would be okay and they would ridicule air mowing.


petemoore

  Boost your 1rst tube Mu.
  Then see if a tube boost [via mu or..]..see what that does...
 
Convention creates following, following creates convention.

km-r

Look at it this way- everyone rags on air guitar here because everyone can play guitar.  If we were on a lawn mower forum, air guitar would be okay and they would ridicule air mowing.

edvard

As far as I know, cascode systems are much like mu-amp FET setups, but they look different.
Notice how they pull signal off of the plate resistor at the top of the stack rather than between the tubes as in a mu-amp setup (which would probably work as well).

Here's some variations of the cascode topology: http://www.glass-ware.com/tubecad/TC_Tube_Circuit_Descriptions.html

Why don't we see cascode designs more often? The story goes that the cascode topology was invented to overcome Miller Effect, the bane of bandwidth in high-gain designs. That all ended with the invention of pentodes so the cascode was "forgotten".
Enter the age of transistors, and suddenly we're back to dealing with Miller Effect, so cascode designs start getting attention again.
Whether that's true or not, it still makes a good story  :icon_wink:

I wonder how this would work with FET's?
*edvard runs off to his breadboard

All children left unattended will be given a mocha and a puppy

earthtonesaudio

I'm a dunce when it comes to tubes, so sorry if this is an obvious question...
What do you do with the voltages of the top tube's cathode relative to it's heater?  Do you float the filament up near the potential of the bottom tube's plate or what?

edvard

If you use something like a 12A*7, the heater is a separate connection, so it's not a problem.
Don't know what you'd do with direct-heated designs.
All children left unattended will be given a mocha and a puppy

frequencycentral

Quote from: earthtonesaudio on July 30, 2009, 12:20:52 PM
I'm a dunce when it comes to tubes, so sorry if this is an obvious question...
What do you do with the voltages of the top tube's cathode relative to it's heater?  Do you float the filament up near the potential of the bottom tube's plate or what?

Valve Wizard says: "Heater considerations: Since the cathode of the upper triode is at a high voltage, it may be necessary to elevate the heater supply to avoid exceeding the maximum rated heater to cathode voltage."
http://www.frequencycentral.co.uk/

Questo è il fiore del partigiano morto per la libertà!

earthtonesaudio

Ah that makes sense... but perhaps in starved-plate context it becomes less important.

frequencycentral

Quote from: earthtonesaudio on July 30, 2009, 01:43:26 PM
Ah that makes sense... but perhaps in starved-plate context it becomes less important.

Yeah, my thoughts exactly. I'll breadboard a starved plate cascode device soon, just got to get a couple of other ideas out of my system first.
http://www.frequencycentral.co.uk/

Questo è il fiore del partigiano morto per la libertà!

maarten


frequencycentral

Quote from: maarten on August 01, 2009, 03:44:40 PM
Hello,
Did you see: http://www.jogis-roehrenbude.de/Leserbriefe/Hermann-Mini-Amp/Jeschkeit-Mini-Amp2.jpg ? Preamp is a cascode of 2 submini 5840, followed by a submini 5902 output tube.
Maarten



No I hadn't seen that, and I've got all those tubes in stock, so I'll have to try it!
http://www.frequencycentral.co.uk/

Questo è il fiore del partigiano morto per la libertà!

brett

Hi
I'm no expert about cascodes, but I believe that the "top" valve is an active load and not doing much in terms of the sound, so you might as well replace it with a solid state active load.  (ie like the JFET active load in the mu-amp and the BJT active load in the Jordan Bosstone).  Personally, I'd go for the JFET, which is dead simple to configure as a constant current source that will lift gain dramatically (see "foolin' with FETs" at geofex.com for the solid state version). 

This setup might also help overcome the issue of low transconductance in "starved plate" valves.  I've always thought that low voltage valves would be best driving a several Mohm JFET or op-amp stage, rather than a pot (even a 1M pot). 
cheers
Brett Robinson
Let a hundred flowers bloom, let a hundred schools of thought contend. (Mao Zedong)

frequencycentral

That's interesting Brett. I've already done something similar with my modded Pentadriver circuit: http://www.diystompboxes.com/smfforum/index.php?topic=70533.80 Though I hadn't made the link between cascode and constant current source. Now that I look at it again I see that it looks very similar to what you are describing. I haven't tried it with FETs, though I know intricatecode did a similar mod to his Valvecaster: http://www.diystompboxes.com/smfforum/index.php?topic=63479.msg595904#msg595904

I think I'm having one of those 'ah, i see now' moments - the top valve ina cascose is a CCS right? So much to learn.............
http://www.frequencycentral.co.uk/

Questo è il fiore del partigiano morto per la libertà!

intricatecode

is it really working like a CCS? I admit that  i don't know enough about these things to have a learned opinion but just by looking at the circuit I see some diferences in topology between the jfet CCS and the cascode. Might as well go and try the cascode and compare it with the jfet CCS and listen to the diferences and why not how about a cascode with a jfet CCS on the top valve instead of the resistor?

frequencycentral

Yes there are differences, it was very early when I typed that and I was still breary eyed - I've woke up now. The cascode output is the plate of the top tube, the CCS output is the plate of the bottom tube.

Quote from: km-r on July 29, 2009, 09:27:42 PM
Quote from: Ripthorn on July 29, 2009, 09:08:13 PM
Rick, I expect you will also try a dual cascode setup (cascaded cascodes).  That would mean uber gain.  Let us know how it works, seeing as how time to actual sit at the  bench is almost non-existent right now.

TUBE BSIAB2?

A tube BSIAB2 could use FETs for the top part as Brett suggests, and a pentode or triode for the final stage - if it was a triode from a dual triode package that would leave a triode spare to make a pre-boost. Hmmm......
http://www.frequencycentral.co.uk/

Questo è il fiore del partigiano morto per la libertà!