5842 grounded grid

[alexg]

Tube newbie question: I had this 5842 grounded grid circuit built from Yeo's diyparadise.com, what is the best way to bump the output a little bit?



Instead of the 120H choke, I used a 10K resistor and the cathode resistor is 160 ohms.

[DougL]

Maximum gain of the stage will be obtained when you use either a choke or a constant current device.

Short answer, use the choke specified.

Good luck.

Doug

[Eli Duttman]

The O/P impedance of a GG stage is HIGH. As the load impedance approaches infinity, the gain approaches the mu of the tube. A 10 KOhm resistor is COMPLETELY inadequate as a GG load. A good constant current source (CCS) is a reasonable approximation of an infinite load. To keep the gain up, DC couple a voltage follower to the 5842's anode.



edit: fixed typo

[alexg]

I am a complete newbie on this, can somebody please give me a schematic of a CCS circuit that I can use?

I have looked at chokes locally available and I have to contact some transformer winder and see if they can make the plate choke.

Thanks.

[Sheldon]

Here's an example using a current IC:

http://www.pmillett.com/current_source.htm

And here's a bjt version. The pnp version could be used as a plate load:

http://www.diyaudio.com/forums/showt...231#post829231

[JoshK]

Sorry for drudging up an old thread, but I had some more ideas/thought experiments/musings/questions.... GG is the right way to do I/V for a current output DAC using a tube as far as I have learned, so how to get around some of the issues involved are my questions.

As one poster said, the GG has a high O/P Z, surely this isn't great for driving the output directly then. Also it could use the assistance of a CCS as a load to keep gain up. How about doing the equivalent of a mu stage with the bottom triode a GG? Would it work?

So you are driving the cathode with the DAC, grounding the grid, and taking the plate to the grid of a triode on top of the gg tube. There is a resistor inbetween the plate of the lower tube and the cathode of the upper and you take the output from the top of the resistor. Would this give you the low O/P Z of the CF with the benefits of the GG stage (low I/P Z for the Idac)?

P.S. I realize you could also do a GG cascaded into a CF stage, but would you need to cap couple?

[Eli Duttman]

Josh,

The 5842 has good drive capability. A maximized 5842 GG stage needs CCS loading and voltage follower buffering. DC couple an IRFBC20 MOSFET source follower, as described in MOSFET Follies, to the triode's plate.

IRFBC20 data sheet here.

BTW, just because it's GG doesn't mean you can forget precautions against parasitic oscillation. You have to put RF chokes in the heater power lines at the socket.

[DougL]

Sorry for drudging up an old thread,

That's the norm for this board.

As one poster said, the GG has a high O/P Z, surely this isn't great for driving the output directly then. Also it could use the assistance of a CCS as a load to keep gain up. How about doing the equivalent of a mu stage with the bottom triode a GG? Would it work?

High output Z is a relative statement. Because the cathode isn't bypassed, and a high value plate resistor is used, output Z is much higher than input Z. Your Idea with the Mu follower is sound, By the way.

Would this give you the low O/P Z of the CF with the benefits of the GG stage (low I/P Z for the Idac)?

Potentially, yes.

For a current DAC, I tried a 1:4 step up using Edcor transformers and a resistor. For simplicity it's killer. Very pleased with the sound. Total cost $20 Edcore 600:10K transformer There is a more appropriate one, but this is what I had from another project. Here is the link for the transformer output.Raleigh Audio RAKK DAC passive output page.

HTH

Doug

[JoshK]

Thanks Doug & Eli! As I looked through more examples on glass-ware, I noticed that essentially broskie was doing this with an added CF in front in some of his circuits. This also gave me a little more confidence that its worth a try.

Eli, I will certain be very cautious with oscillation. I have read a lot of tips on how to keep a handle on this. However, I wasn't sure what kind of ferrite beads were appropriate? The ones that have leads like a resistor, or the clamp on ones?

[Sheldon]

Originally posted by JoshK
Would this give you the low O/P Z of the CF with the benefits of the GG stage (low I/P Z for the Idac)?

I can understand why low input Z would rule out a GG stage for some applications, but I'm not sure why low input Z would recommend a GG stage over high input Z stages. It's easy to lower input Z with a shunt resistor, if the preceding stage likes a low Z load, no?

Sheldon