• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Folded/shunt cascode differential question

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
You need a protection zener for the lower device if J112. A 15v one should work well

The IXTP08n100d as well as the other higher current versions from IXYS have a higher VGS value for lower current (<50mA) which will allow the J112 to operate at higher VDS so capacitances at play are reduced. Overall FR of the stage is improved at HF this way.
Btw: and they sound better!
Ale
 
Updates! Board grew a bit, but I was able to route everything much more cleanly thanks to the ammo pack suggestion and (ironically) adding a few more/bigger parts. Now about 3.75" by 2.25" and accommodates a standard 1" spaced heatsink for the higher dissipation TO220.

schematic-cascode-cascode.png


board-2nd-ed.png


Gotta say, laying out boards is kind of fun (in the way that ptp wire routing OCD is fun).
 
Hard to believe that I've had boards for this on hand now for a few months. I am finally working my way back around to the project and implementation.

I'm wondering about start-up behavior. I have the option of powering this with a tube rectifier, which would delay and slowly ramp up the B+ (thinking 5AR4). This seems like a good idea because it will allow the tube to start conducting before HT is applied to the anode of the tube and emitter of the BJT through the shared CCS. Lacking that more controlled start up, I worry the BJT current will only be limited by the CCS with the tube heaters cold.

Anyone care to confirm this line of thinking?
 
I'm wondering about start-up behavior. I have the option of powering this with a tube rectifier, which would delay and slowly ramp up the B+ (thinking 5AR4). This seems like a good idea because it will allow the tube to start conducting before HT is applied to the anode of the tube and emitter of the BJT through the shared CCS. Lacking that more controlled start up, I worry the BJT current will only be limited by the CCS with the tube heaters cold.

When you build the Shunt Cascode board, start with the CCS adjusted to the low end of the expected range. When it's warmed up, adjust the CCS until the output is at ½ the total output swing. The swing is almost equal to the anode voltage of the triode.

The maximum current in the output is 2x the idle current, because the output voltage is then about equal to the anode voltage. The PNP does not allow the voltage to raise any higher, so that is a stable condition. The PNP dissipation is very low at this point, because Vce is very low.

So long as the output resistor is well-rated for the power, nothing bad will happen, even if it is left running in this state for hours.
 
I would suggest to return output through servo RC to R2, to stabilize the working point. The time constant should be far from R1/C1 to avoid possible instability.

I always plan to add a servo to my Shunt Cascode PCB.

But in practice, regulating the heater supply is usually sufficient, in my high-gain driver design. Of course, it depends on the quality of the valve, but I can say that the 6Э5П with OTК rhombus marking has very stable anode current, over a long operating time.

The last lot of 6Э5П I got from Ukraine is marked ПЕРЕПРОВ, so we will see if they are good enough.
 
So long as the output resistor is well-rated for the power, nothing bad will happen, even if it is left running in this state for hours.

Thanks for this and the adjustment tips, Rod.

Of course the emitter is setting the anode voltage and the collector resistor is dropping voltage, effectively limiting current. It shouldn't matter if the tube is an open circuit. That makes perfect sense.

I let this project get a little dusty while I was distracted by other things. I came back and have to relearn it apparently. I'm also easily confused by PNP :)

Pics of the board:

shunt-2.jpeg


shunt-1.jpeg
 
Member
Joined 2008
Paid Member
Interesting to see your project SodaC. I was just telling Rod of my plans to use his shunt cascode circuit as the basis of an active line level XO and that one of the approaches I was looking at was in Small's subtractive crossover, which just happens to need a balanced amplifier stage.
Looking forward to hearing how your amp circuit works out. Thanks!
 
Me too :)

I'm planning to test the shunt cascode boards single-ended/unbalanced initially to get close on the trimmer settings and then wire up with the LTP CCS (simple LED and BJT cascode).

It's a bit fiddly with all the trimmers (voltage reference, anode CCS, tail CCS). I'm hoping things are fairly stable once dialed in. Provided this works out, I'm very interested in trying the shunt cascode single-ended for a phono pre as well.
 
I just built and tested a folded cascode differential based on PRR's post #5 in this thread, intended for use as a phase splitter. I needed something with decent gain that could run off the low supply voltage available (only 155V, output stage is hybrid). Triodes had to be from an ECL85 as I'm using the pentode in the output stage.

See attached. Supply is from a simple Maida regulator, voltage source for the PNP bases is a resistor divider bypassed to B+ (helps with PSRR), anode current sources are just resistors. Gain is about 45 (~90 each side, halved due to grounding the second input). No issues with oscillation or instability, can swing >100V peak to peak, and perfect balance. On start-up the outputs sit at ~124V until the tubes warm up, then falls to the normal operating point with some undershoot.

All in all a very useful building block for the right circumstances.
 

Attachments

  • Folded cascode differential phase splitter.JPG
    Folded cascode differential phase splitter.JPG
    306.4 KB · Views: 520
Finally in the testing stages:

img_20190112_150247764-01.jpeg


I'm troubleshooting at the moment. Without the tube I can set the voltage reference and output level no problem. With the tube installed, the reference ends up around 50V (target 150V) and the output at 0V (target 75V).

I'm playing with the tail CCS to see if that's an issue. I'm thinking I probably killed the ZTXs though. At one point I accidentally cranked the anode CCSs the wrong way and was alerted by a dropping resistor heating up.
 
This was also on my mind. I plan to lower the tail CCS current and try again slowly increasing the anode CCS to see if it's the issue.

I don't understand why it's affecting the voltage reference and that's what got me questioning whether I've killed something.

Rough schematic attached.
 

Attachments

  • Schematic_PP-shunt-cascode-monoblock.pdf
    47.7 KB · Views: 198
It's possible something is dead, but if the tube eats the whole CCS output, and is still hungry, the output will fall even if the reference circuit is good. I would say this is the most likely problem.

Don't worry about having a little too much current - if the output resistors are correctly chosen, the power dissipation is limited.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.