Preamp working but not

No information transferred. Post a schematic.

...Jim Thompson

--
| James E.Thompson, P.E.                           |    mens     |
| Analog Innovations, Inc.                         |     et      |

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

Why? The circuit is the standard class a tube amp circuit. I've shown that it is amplifying the input signal. So it is only on the output stage that is the problem? (I don't see how it can be anything else... besides there is no schematic but any search on google for class a pre amp should show the idea. The operating conditions are fine as I get the proper Q point I wanted)

---||---+---R1---- To mixer | R | G

I've tried with R, R1 and removed one or the other... all with same results.

R1, and R at 100k, R at 700k, etc..

If R = 100k then my amplification factor is only 30 so there is some loading going on. R=700k gives about 90x.

So it would seem that the mixer has a smaller input impedence of 100k? (seems to be about 1 to 2 amp factor so thsi would imply a very small imput impedence?)

Again, The circuit seems to be working except for output "loading". If you could explain to me how the circuit could work fine yet not work fine, excluding the output loading issue, then I'd be willing to draw up the circuit. (not that I'm saying your wrong but I'm convinced it is a loading issue but in some sense this is a bit weak unless mixers generally have a lower input resistance than I am assuming)

i.e., my scope is able to get the voltage without bringing it down but the mixer and soundcard are not. This tells me that it is a loading issue(if the mixer used the same circuit as the scope then it should get ~500mV peak instead of 5mV peak.. and hence be louder?).

Clearly you've bungled something, but since you know it all I guess you don't need any help.

...Jim Thompson

--
| James E.Thompson, P.E.                           |    mens     |
| Analog Innovations, Inc.                         |     et      |

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

Well, since clearly you think you know it all you could tell me why. At least I explained my logic but I never said I was right.

Aren't you the one that goes around calling everyone weenies? Sounds like your being the weenie now! I'd expect obama not to explain his logic but not the same from you.

You wanted a schematic and there isn't one. I told you that it is the standard class A single ended amp that is in just about every tube book that exists. You say I bungled something in my circuit... maybe, but if you can't troubleshoot the results then it doesn't hold water.

If you can't explain a reason why my logic about it being on the output side as being flawed(any counter example will do) then I can't take the time to draw up a circuit.

If this was more complex circuit than just a few resistors and capacitors(4 resistors and 3 caps to be exact) then I might have posted it... but if you think I'm that incompetent that I can't wire up a few resistors and capacitors then screw you... don't bother to try and help me because I'm beyond help then. If you think that I might have wired up something wrong giving the problems then state a possible reason why my logic is wrong and then I'll be willing to investigate the other side.

Don't pull an obama and claim that you know more about my circuit than I do without ever seeing it and don't say I'm a moron without any proof. It's a sad state in this group where so called educated engineers can only show there ego's.

Now, if your telepathic or a seer then by all means say what you want about my circuit... but if that is the case then why did you ask in the first place? If you swear that it is my circuit design, then that is fine... but why not at least trust me on the initial assumption and try to prove it fallacious or solve the problem instead of just assuming I'm a moron. (Which in this group people love to assume about others)

If I'm a weenie, then you're a wuss.

Engineering is not subject to opinion.

How can you troubleshoot an undefined circuit?

Don't bother me anymore if you "can't take the time to draw up a circuit".

I suspect you nailed it! Why are you so defensive? Are you a known bungler ?:-)

I AM telepathic. I see a bungler ;-)

...Jim Thompson

--
| James E.Thompson, P.E.                           |    mens     |
| Analog Innovations, Inc.                         |     et      |

Have no idea what the output impedance of your tube preamp is but often it is as much as 10K-20K-30K and sometimes as much as 100K ohms.

Many "modern" pieces of equipment (your mixer?) have line-level input impedances circa 10K ohms, sometimes as low as a few K ohms. Most solid-state components have output impedances in the tens-of-ohms-to- low-hundreds-of-ohms so they don't notice the loading of the next input.

You can actually measure the input impedance of your components with a signal generator, a few substitution resistor, and AC voltmeter or scope.

Tim.

All elements inside the tube are the same. Think about it. They are metal. They are inside a glass envelope. There is a vacuum inside. I suggest you put the high voltage on the cathode, the signal on the plate, and get your output on a grid. Any grid, they're all the same. One day I'll show you how to get 1W of signal power from the filament. On a cold cathode tube.

Maybe it's receiving "I Love Lucy"?

It's a William "Fred" Frawley design ;-)

...Jim Thompson

--
| James E.Thompson, P.E.                           |    mens     |
| Analog Innovations, Inc.                         |     et      |

(4

ou

Thing is, those resistors (and capacitors) as well as the configuration of the active device set the impedances in the circuit.

ong

d

We've got two impedance mismatches here:

1. The one between your preamp and your mixer.
2. The one between you and the real world that makes you think that it has to be either the input side or the output side and seems to be occupying most of this thread.

Tim.

What in the world is R1 doing? Unless the mixer input is AC coupled, you don't need R either. What is the value of the plate load resistor? Unless the cable between the amp and mixer is around a foot or so, your setup will have lousy high frequency response because of the cable capacitance. You don't drive cable from a high impedance source. You need an (ahem) cathode follower; that will fix your gain problem also. Do you have any particular prejudice against opamps?

Tam

That's better than a Fred Bloggs design.

--
http://improve-usenet.org/index.html

aioe.org, Goggle Groups, and Web TV users must request to be white

not

Who knows? I haven't seen any, or any replies to him.

--
http://improve-usenet.org/index.html

aioe.org, Goggle Groups, and Web TV users must request to be white

A very complex schematic above of the tube amp circuit. It might take a few weeks to go over it.

It follows the the standard Class A single ended tube topology as I originally said(so anyone in the "know" shouldn't need it).

I have tried direct input coupling and it doesn't make a difference. I've tried different values of output coupling and I've tried non-polar electrolytic caps(two hooked up back to back). They are all the correct voltages(800V combined).

The DC point was chosen to give the least distortion(although with such a small voltage input swing it shouldn't matter too much). (load line intersecs Ec curves at right angles which in theory should give the lowest distortion and well below the max power dissipation spec)

The data sheet is at

My Dc point is at Ec = -2.25V, Ebb(B+) = 450V, V_DC = 225V, I_DC = 0.5V. All check out with the scope/meter so I know the tube is operating as it should.

Hence 99% chance the problem is on the output stage. With a very low load I get proper scope results(as I have already said) so this has to be a loading issue(impedence mismatch). (i.e., the tube part of the circuit is really not part of the problem like a few seem to think... but I guess there's always the chance)

Now to the output stage where most likely the problem lies: The tube's plate resistance at 250V is about 85k.

I've used various resistors in parallel with the mixer and ran it direct and such... I can't see any way that if the mixer has infinite input resistance how it could competely load the amplifier.

Again, everything works on the scope except I do have the issue of the large dc(several volts) which I guess is due to the charging up of the output coupling cap and leakage(was bad when I was using 150uF electrolytic caps. After a while it does drop to a few mV with 1uF cap. (I could decrease the capacitance but I wanted to get a pretty low freq response)

I've tried hooking up the output up directly to the mixer, through a series resistor, through a parallel resistor and with both. Nothing works. The freq response is fine so it's not being attenuated due to the bandwidth.

The signal into the amp is slightly stronger. I do not know how much but on the db meter of the meter the original signal stays at the 20db led and the

*amplified* signal sometimes goes into the 0db led(which is the next one).

So it is the case that it is being amplified. It is possible that my tube circuit is not amplifying as expected but the scope is showing about 90x amplification. I can't imagine the tube somehow crapping out when I hook up the mixer unless it is somehow loading it yet not crapping out on the scope? (except of course because of loading issues which isn't the tube circuits fault but the output stage)

In anycase I'm probably going to wire up the power stage and see if it gets loaded too(shouldn't as the gate has virtually infinite resistance and I know with about 500k I get 90x amplification(100k gives approximately 30x)).

Have no idea what the output impedance of your tube preamp is but often it is as much as 10K-20K-30K and sometimes as much as 100K ohms.

Many "modern" pieces of equipment (your mixer?) have line-level input impedances circa 10K ohms, sometimes as low as a few K ohms. Most solid-state components have output impedances in the tens-of-ohms-to- low-hundreds-of-ohms so they don't notice the loading of the next input.

You can actually measure the input impedance of your components with a signal generator, a few substitution resistor, and AC voltmeter or scope.

Tim.

------

Then that is most likely it. If the mixer uses only 10k ohms input impedence then that will definitely load the amp too much(needs to be about 500k for

90x... 100k drops that down to 30).

I know the signal is being amplified but maybe just 1.2x. So that would explain it.

I always thought that the input's to amplifiers were always extremely high... just thought that was the way it was done so as to not overload the previous stage. I guess that was the misconception I had.

So if thats the case then my circuit is working fine and I just either need to match the impedence(I do have some small output transformers that might work) or go ahead and wire up the power amp stage and work with it(still same problem of course but I can then drive a speaker directly hopefully).

Or, I guess I could make a simple sold state amp using an op amp and drive the amplifier with that if I wanted?

In anycase if what you have said is true(and not that I doubt it unless someone else contradicts it) then there are no major problems and I can move on to the next stage.

I don't have a signal generator but I used an ohm meter to try and measure it and it seems to be capacitively coupled becaus the "resistance" kept increasing(from 1M to 5M at least). Hopefully I'll be able to either built one(using a pic) or buy one sometime soon(need it for other reasons too).

Thanks, Jon

es

e

I don't understand "nothing works". Do you expect that increasing the output impedance of your preamp (resistor in series) or decreasing the input impedance of the mixer seen by the preamp (resistor in parallel) will increase system gain?

Do you think that resistors have gain?

You might want to google "preamp" with "cathode follower" to learn how others solved your problem.

Tim.

he

"Extremely high" is relative.

Compared to an op-amp stage's output impedance, 10K is really high.

Compared to a tube preamp output with a 470K plate resistor, 10K is really low.

Look at the preamp circuits in the RCA RC-30 tube manual... every preamp specifies the minimum input impedance of the next stage. And most use cathode followers.

Tim.

Nowhere in this original posting did you mention that the plate load resistor was 450K (BTW, where did you buy that value ?:-)

Your mixer isn't "overloading" the preamp, it's simply obeying resistance fundamentals.

...Jim Thompson

--
| James E.Thompson, P.E.                           |    mens     |
| Analog Innovations, Inc.                         |     et      |

Jon,

the problem is that the input impedance of your mixer is much less than the output impedance of your tube amp. Think about what determines those two values.

Mark

A couple of seconds, more like.

This tube, running at this tiny current, with that huge plate load resistor, will have very high Zout and rotten hf frequency response.

That's about it.

John