[cws5]

I've begun to try to better understand the circuits in my amps. I've had success with the MG Head. It's got two plate-follower stages that look much like those you'll find described at the various tube info sites on the web. But the Darkvoice 332 is not so easy to figure out. I'm working now on just the power stage.
 
Here's a description of the circuit (only one channel), which is some kind of cathode follower: the driver stage 6AK5's screen grid (it's a pentode run in pentode mode) feeds the power stage triode's grid. There's no resistor on that grid. B+ hits the plate. The signal comes off the cathode, then splits in two, each line hitting a cap. There is no resistor between the cathode and either of those caps (unlike most cathode follower circuits I've seen on the web. One cap's signal is the positive AC signal, and it's wired to one of the phone jack pins. The other cap sends the AC signal to ground, with that ground wired up to the phones jack ground, plus a 10kohm resistor attached from that grounding point to the terminal of the phones jack where that positive AC signal came in. 
 
So I'm mainly trying to understand why, unlike most cathode follower schematics I see, the cathode load resistor is AFTER the cap. What's that cap doing there?
 
I'm also curious why, unlike most schematics I've seen, there's no resistor preceding the power tube's grid.
 
I'd be grateful for a little enlightenment here.

 

[nikongod]

Regarding the uncommon pentode setup:
Someone else will have to comment on that. Im not 100% sure how this works, but I have some lucky guesses
 
In many ways triodes are funny devices (this extends to pentodes as well but they are even funnier). Most people think of them having 1 input and 2 outputs. The control grid as input, and the plate and cathode as outputs but in fact if you do some unconventional things ANY part of the tube can be used as an input or an output. Crazy, hunh? 
 
Anyways, to answer the other part of the question the DV332 has "proper" cathode resistors.
 
In this photo of the 332 I found:
 

 
The cathode resistors can be seen on the top right. They are green cylinders with a bolt going through them.
 
As for the resistors after the output caps, they are there to discharge the caps. When the amp is first turned on (or if it is idled for a long time with no headphones plugged in) it is quite possible for voltage to leak across the output caps and be stored on the output side. If these resistors were not present this causes a VERY loud thump/pop when the headphones are plugged in. Depending on your lack of luck it may possibly be enough to damage the headphones - the resistors reduce/eliminate this effect.  

 

[cws5]

Many thanks, nikongod. You've solved some mysteries for me.
 
I had assumed from the size of those green casings indicated that they were capacitors, which was pretty easy to do since there's no marking visible on them (it must be on their underside). thanks for letting me know that they're resistors. I've never seen resistors of that design (thru-bolt with one terminal perpendicular and another axial). Maybe you know where might I see an image of such resistors on the web?
 
I understand your explanation of the resistors after the coupling caps. Thank you.
 
 
 

 

[nikongod]

I hope this link works. If not your on your own

 
link to mouser site. Many similar resistors.

 

[cws5]

Thanks again, nikongod. I tracked down the resistors. They are Chinese -- SX Electronics vitreous enamel high power wire-wounds.
 
And an update: I was wrong in believing that the driver-state pentodes are run in pentode mode. they're not. They're triode-strapped. I hadn't noticed originally but upon closer inspection found that the plate and screen-grid pins have been connected. This makes sense, as I originally couldn't figure out how the plate would yield an output if only the screen-grid was B+ powered. In that post of mine that you replied to a few months or so ago you helpfully suggested that I look to the cathode-follower as the main cause of the difference in sound between the MG Head and the DV332. You sure were right, as the DV332 isn't a pentode driver amp. It's a triode-strapped pentode driver amp.
 
Also interesting about the DV332: it's interstage direct-coupled. Not many amps at this price point are direct-coupled, so that's pretty unique. 73v DC at both the plate of the driver and the grid of the power tube, with no coupling cap. 
 
I've measured all the voltages, recorded all the resistor values, and sketched the amp's schematic. So I'm now going to try to work through the numbers using the tube data sheets to try to determine why the amp is so lacking in the midrange. Listening to the amp, the frequency curve seems to be like what some call 'the smile' -- heavy on both the low and high end, with a pronounced dip in the middle.  If that could be somehow fixed by changing resistor values or adding cathode bypass caps or whatever, then the amp would be a really good one. As it stands, it's an ear-bleed inducer; turning the volume high enough for sufficient midrange and the high end is like knives. Let me know if you've got any ideas about this.

 

[galyons]

Quote:

I've measured all the voltages, recorded all the resistor values, and sketched the amp's schematic. So I'm now going to try to work through the numbers using the tube data sheets to try to determine why the amp is so lacking in the midrange. Listening to the amp, the frequency curve seems to be like what some call 'the smile' -- heavy on both the low and high end, with a pronounced dip in the middle.  If that could be somehow fixed by changing resistor values or adding cathode bypass caps or whatever, then the amp would be a really good one. As it stands, it's an ear-bleed inducer; turning the volume high enough for sufficient midrange and the high end is like knives. Let me know if you've got any ideas about this.

Hi,
Would you be willing to share?
Thanks,
Geary
 

 

[galyons]

What are the values of these resistors?
Thanks,
Geary
 
Quote:

Thanks again, nikongod. I tracked down the resistors. They are Chinese -- SX Electronics vitreous enamel high power wire-wounds.
 
A