• 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.

Tube Theory

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi,

I'm building pre-amp consisting of 3 gain stage of paralleled ECC81. As this is quite expensive, I don't want to get an unexpected result by matching it to the wrong amplifier (Aleph).

Too high an output impedance is what I'm worry about. So I need a hint regarding tricks to lower the output impedance, or good site where I can probably learn how to do it (May be I need to replace the last stage with 12AX7?, may be I need to create active load on the last tube's chatode?)

Thanks
 
Ways to lower output Z:

1. Use a cathode follower as the last stage. A CF can, religious protestations aside, be designed to give extremely low distortion and high bandwidth.

2. Use a different tube. Something in the 6DJ8 family, for example, run at 10 ma will have a much lower output Z than anything made from an 12AT/AX7. There are a lot of other choices out there with lower rp and higher current than the tubes you're proposing.

3. Use voltage feedback from the last stage. This is the trickiest to implement, but some intelligent choices and possibly mixing this method with the first two methods will get you superb results.
 
Thanks for the replies…

I have read the cad journal, and I have now simple questions before proceeding with in depth theory and calculation…

Because adding a bypass capacitor in the cathode resistor seems to simplify the choice of the tube and the corresponding bias elements, is it common to use this approach in the pre-amp output? I mean I want to assume that the bypass capacitor is required/mandatory.

Before replacing the tube, do I have to base the calculation on the existing current or voltage gain? (I mean I will first calculate the required gain, which I assume is the gain in the original circuit?) What is the optimum or correct gain, anyway, considering the amplifier input gain stage?

And SY, what is the best candidate for the tube in the last stage? (I want a high-end tube). But I think it will be easy, may be a must, that it is also a dual 9 pin triode.

Thanks again...
 
diyAudio Senior Member
Joined 2002
Having A Bypass.

Hi,

I mean I want to assume that the bypass capacitor is required/mandatory.

No, it is not mandatory.
A bypass cap shunts the AC signal appearing at the cathode to ground thereby optimising the gain in that stage and lowering Zo in the process ( I simplify).

Ommitting the bypass cap under the same circumstances introduces some degenerative feedback for that stage, hence reducing the gain and upping the output Z of that stage.

Either way, you'd better do the maths with the bypass present or not and than design what you prefer, not the other way around.

One question I have for you: is this a line preamp?
If so must you really use the highish gain 12AT7 series of tubes?

Last remark:

Paralelling sections of dual triodes may be beneficial in some applications but if you can avoid it in a preamp, please do.

Cheers,;)
 
My $0.02

A 6DJ8 (6922) as line stage connected as a comon cathode(unbaypassed cathode) it will drive an Aleph front end (around 10K) but marginaly with a 10 mA bias, you will notice a loose bass and not to good highs.

I used this setup for about a year, not bad.

As you are constructing this preamp you are better off selecting an even more powerful tube and check if you can do without a gain stage, selecting instead a CF as suggested in previous postings.

:bigeyes:
 
Causing trouble.....

Sch3mat1c said:
So...why is everyone averse to a cathode follower? This is only one of the things it was born to do, after all...

Tim
Tim,

They're more fashionable now on the forum than they were.
Last year I was actually cautioned by 2 moderators for suggesting that they were usefull! After all, one must not propagate incorrect information :cannotbe:

Cheers,
 
Re: Causing trouble.....

dhaen said:
...They're more fashionable now on the forum than they were.
Last year I was actually cautioned by 2 moderators for suggesting that they were usefull!

This is true! :eek:
I think there have been some solid arguments for them made since then, and some others have experimentally seen for themselves that the anti-CF league is not based on facts.

Joel
(who always thought a CF doesn't sound like anything)
 
Well-reasoned, Brett.

Anyway, Jay, the 12AT7 as a cathode follower will work OK as long as you're not doing anything weird with the load (like driving 50 foot high capacitance cables). A 6DJ8-type (there's a whole family of these) will work even better as a CF, being able to drive higher capacitances without loading down.
 
So there is a pro and contra about Cathode Follower? Many don’t like CF (that’s the fact). I’ll call that a PHENOMENON. It coincidentally happened to me that from (only) a few amplifiers I know (I just realized now), CF designs are indeed “bad” sounding…

Hmmm… I think I will temporarily join the anti-CF league... For safety reason :cool:

Oh yeah, Frank, about the preamp, it was a line stage. This is I think a typical JADIS preamp, utilizing 12AU7/12AT7/12AX7. I wonder why they had survived…:rolleyes: :rolleyes: :rolleyes:

Why should I avoid the paralleled triode? Like in SS, the effect is for tighter bass (beside longer tube life). In SS, there is no side effect to the mid, in tube it seems like there is a terrible trade-off… My option is then to parallel OR to inactivate half the tube. Sound-wise, which one is better?

BTW, SY, I’m not going with the CF. Is that badly-reasoned? ;)
 
diyAudio Senior Member
Joined 2002
NOT ALL CFS ARE CREATED EQUAL.

Hi,

Jay,

Do you mean something like this?

It's a buffered input (CF), followed by an anode follower using a 12AX7 RC coupled into another 12AU7A CF.

I don't know who thought this up but this one hell of a complicated way to make a simple line preamp.

The whole thing can be reduced to a single 12AU7A (or better) per channel.

Cheers,;)
 

Attachments

  • jadis.gif
    jadis.gif
    6.9 KB · Views: 390
Let's just say that there's an awfully large set of classic, highly regarded gear that uses a cathode follower. And that no-one has ever presented any measurable or audible evidence (beyond bald assertion) that CFs as a topology are inherently flawed for their intended application. Yes, you can do a bad CF design. That puts it in the same category as every other topology.

The acceptance or rejection of CFs is more of a sociological than electronic issue. If that's what's important to you, than by all means, don't use it.
 
As for audible evidence, what exactly are you looking to be presented? Twice, once about 10 years ago, and again about 3, I built a number of CFs, switchable bypassed (level matched, good earthing and PSU) into two power amps or my ES headphone driver. The measurements between the two topologies was slight. The CF was plainly audible, some variations more than others. I've also tried a number of other times less formal experiments, into various loads, some variations including CCS as cathode loads. The best so far has been the Vacuumstate circuit I'm currently using, which drives my TVC a little better than some of the CC I've tried. Directly into, say, a high input Z poweramp, I would probably choose something else. I've built about 10 different linestages this year to try them out, before I reached this conclusion. Soon, I'd like to try some pentode CF's again, and some mu stages, and some Whites and even a Broskie, just to see. The experimentation can be fun.

There are classic designs out there that are well regarded amongst some people using CF's. In a commersial design, there is a requirement to be able to drive a wide range of poweramps, so some designers choose them. We are not so restricted in diy to the came commercial considerations. Having owned and/or used a number of commercial, 'well regarded' preamps, my feeling is, well regarded by whom?
 
Jay,

I take it that your Aleph is hard to drive? If you don't want a CF, try a ECC99 anode follower, or, if this isn't enough gain, the before mentioned 6C45. I don't know why you'd need to complicate things with three stages...unless the Aleph is hard to drive and an insensitive beast.

An EC99 or 6SN7 as a VA and CF stage is simple, sounds fine, and will drive most amps. A simple circuit(even one using a CF) will generally sound better than a circuit with unneccesary additional stages.
 
Joel said:
Brett is siding with the Frank D. philosophy of DIY which is that any circuit that can be used in multiple applications is "bad". A "good" circuit is specialized to a single application.
I think that is bollocks.

I have no idea where you got that from. It's not what I said nor what I meant.

You should write for politicians.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.