• 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.

Can someone give some info on this tube circuit please?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
His circuit uses higher B+ (200V vs 140V). Try 20k first, that should work. The plate current will increase by the reciprocal factor, as the grid voltage is constant. Is the power supply just passive dropping resistors? One or more of those may have to be decreased.

The input section has low plate current, another limitation of this circuit.

Here's the whole thing.
Thanks.
 

Attachments

  • 003.jpg
    003.jpg
    226.8 KB · Views: 221
how will you know better when you have never seen the start...?

you have to start somewhere and decide for yourself if what you heard is good or not good to YOUR EARS......not our ears as we are not listening to your finished amp.....

if you assume from the start that such a circuit can be better, then you do not learn anything...you have no reference, just the say so of others...

a cathode follower in that amp has low enough output resistance, 68k loads will be peanuts...

a 12au7 cathode current at around 2mA, gives transconductance of about 322 ohms if my calculations are correct...others can correct me.....http://www.mif.pg.gda.pl/homepages/frank/sheets/093/1/12AU7A.pdf
 
"how will you know better when you have never seen the start...?"

I hoping tube gurus may see obvious problems that are measurable rather than just relying on listening.
I'm one that like the specs/measurements right first in solid state before drawing subjective conclusions, after all that's how things are designed in the first place by the ones in the know, get the specs/maths and measurements right first then listen. Not the opposite.

" 68k loads will be peanuts..."
Your saying now 68kohms is peanuts, other say it's not great, who am I to believe now? I 'll wait for more info from others to see who backs who as you have just asked "others can correct me."

Cheers George
 
georgehifi said:
So even though it's a couple hundred ohms output impedance, the 33kohm cathode resistor is the controlling factor of what it can and can't drive without any impediments.
Yes and no. The cathode resistor in the CF sets the CF current, which affects the load driving ability. Unfortunately it also affects the distortion. You need to decide what balance you want to strike: low distortion into a high impedance load (high cathode resistor) or not too high distortion into a lower impedance load (low cathode resistor).

As I said, this circuit is hampered by low supply rail voltage and poor valve choice. You can fiddle with component values to pick a different compromise from that chosen by the designer, but if you want better performance you need to address the real problems.

Another possible problem is noise. If this circuit is used in a high gain setup, so the volume control stays at the low end, then thermal noise from the 100k input resistor (about 6uV) might become an issue. If used with high volume control settings then it should be OK.
 
even with op amps, driving 68k load is peanuts when your opamp output impedance is 600 phms..

the 33k at the cathode of the cathode follower is there just to bias the tube or set its operating point, as DF96 pointed out, the output impedance of a cathode follower is its cathode internal resistance, in this case 1/transconductace, i estimated it at 322 ohms which i arrived at by dirty calculations, this is what i mean when i said i could be wrong....but cathode followers have such low output impedance.....
 
Broskie's CCDA works on the premise that the common cathode plate load is the same as the cathode follower resistor and that the plate voltage of the common cathode gain stage is at 1/2 the B+....the reason being the net ac current drawn from the psu cancelled...

so the 100k plate load resistor for the first stage is indeed old fashioned...
make them the same 33k or even 22k and adjust the cathode resistor of the first stage so that you get half the B+ on the plate...

tubes are fun to work with.....
 
As I said, this circuit is hampered by low supply rail voltage and poor valve choice. You can fiddle with component values to pick a different compromise from that chosen by the designer, but if you want better performance you need to address the real problems.

the 6dj8 and its siblings are one such fine tube to use on such low voltage rails...
with the 12au7, i will use upwards of 200, up to 300 volts,
but keeping in mind the plate dissipation ratings for longer tube life...
 
George, this circuit can drive 68K no worries with regards to voltage swing but distortion will be a bit higher as opposed to a 100K load. DF96's observations are spot on.

It is obvious this circuit is designed to drive a tube amplifier only. The output coupling capacitor is too small to give a good output impedance at 20Hz with lessor load impedances. To drive 68K, I would use a 2uf device. Also, the 450K resistor to ground is in parallel with whatever load you place on the circuit; I would change that to 1M since its purpose is to drain the coupling capacitor to 0 volts during warmup.

(I like to have a timing constant that is at 2Hz or less to prevent phase shift at the bottom limit of human hearing; but be careful that this timing constant does not go lower than that of the power supply or the unit may seem a bit unstable)

It really needs a higher B+; if this were my circuit it would be about 275-300Volts. If the voltage were boosted and the cathode resistor R39/R40 tweaked slightly, you could drive a solid state amp quite well but the coupling cap should be a touch larger in that case.

You can reduce the gain by dispensing with the cathode bypass cap, but the circuit will work better in a number of ways if you retain it. However the cathode resistor without the bypass will create degenerative feedback which might be helpful reducing the distortion when driving a lower impedance.

Since you seem to be new at this, if I were you I would assemble the kit and not place any expectations on how its going to work, since a lot will depend on how well executed the construction actually is. If you can get it going that's a good thing- then I would look into a higher end kit. If you really want to do comparisons with your Lightspeed, you will want a circuit that actually shows off what the active circuit can do for a valid comparison.
 
georgehifi,

I would be careful about solid state specifications, as the only way to initially judge a solid state amp. Find out more about the topology, designers and marketeers goals, etc.

1. A solid state amplifier from one company had extremely low distortion into a resistive load. Good specs. It used large amounts of negative feedback to get that low distortion.
Then customers connected those amplifiers to their loudspeakers.
The amplifiers blew up, because they did not have enough phase margin to work with real loudspeakers (reactive impedance, not an ideal load resistor).
This was not a one-off event, All the amplifiers were returned to the factory.

2. A well regarded amplifier from another company had extremely low distortion into a resistive load. Good specs. It used large amounts of negative feedback to get that low distortion.
Then a customer connected that amplifier to his loudspeakers that were made by a famous manufacturer.
It sounded distorted. But even worse, the amp blew up the tweeters.
It did not have enough phase margin to work with those loudspeakers (reactive impedance, not an ideal load resistor).
Then, a lower powered solid state amplifier with higher distortion specifications was connected to the same loudspeakers (with repaired tweeters), and the sound was good, no distortion, and no blown tweeters.

"I am just sayin" Detective Baez, Blue Bloods
 
"The output coupling capacitor is too small to give a good output impedance at 20Hz with lessor load impedances. To drive 68K, I would use a 2uf device."

The .47uf into a 68k load is still -3db at 5hz which is fine by me but if I assemble this thing I was going to up that to 1uf.


"As I said, this circuit is hampered by low supply rail voltage and poor valve choice."

This could be upped as we have 230v here and if setup using the 200v primaries that will give higher B+, but then other things have to be considered as well heaters ect, all too hard, time is better spent making Lightspeeds.

I'll just build this thing as is not a perfect match, do the A/B demos for customers with the other preamps I have against the Lightspeed into 68k load.

Cheers George
 
i can surmise that the reason the designer chose a b+ of 140 volts and a plate load of 100k for the gain stage is that he liked what he heard, perhaps the 2nd harmonics which people claim as responsible for a "fat sounding" amp....

i built my first tube line stage on a b+ of 50 volts and plate load of 4.7k using the russian 6N23 dual triode.....sound was indeed very good to my ears...
 
George, the circuit topology isn't that bad. As others have pointed out, it is set up to cancel the 2nd harmonic. But I would think about the fact that with a few simple changes, you could build a better sounding circuit.

The tube choice is one- if it were me, I'd use a 6SN7 instead of a 12AU7 which would not be hard since the 12AU7 has its roots in the 6SN7 (via the 6CG7 or the like). So the bias/setup points are similar. Some people prefer point to point wiring and as a simple line stage that would not be so hard.

IOW, I wouldn't give up just because the first circuit (or the tenth one you've looked at, who knows) has design problems like this. One thing you might want to think about is a modified Lightspeed style circuit to manage the volume and balance, perhaps even setting it up so that the circuit has no gain, since a lot of your sources don't need it.

After all, one of the aspects of a passive system (as has been pointed out elsewhere on the web) is the lack of a buffer, and this circuit could be modified with only a moderate amount of cut and splice to just be a buffer with no gain.

Good to see you here by the way- a whole ton of useful information on this site.
 
One thing you might want to think about is a modified Lightspeed style circuit to manage the volume and balance, perhaps even setting it up so that the circuit has no gain, since a lot of your sources don't need it.

Already done that With Allen Wright's (rip) SLCF (Super Linear Cathode Follower), I built both into in a Lightspeed, and it wasn't nearly as as good as without them.

I've posted the A/B up here ages ago, probably in the Lightpseed Attenuator thread.

Lightspeed Attenuator a new passive preamp

Cheers George
 

Attachments

  • Super Linear Cathode Follower Improved .JPG
    Super Linear Cathode Follower Improved .JPG
    31.5 KB · Views: 162
  • SLCF Super Linear Cathode Follower.jpg
    SLCF Super Linear Cathode Follower.jpg
    31.3 KB · Views: 163
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.