• 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.

first go at design - pp

Status
Not open for further replies.
Jump to Latest
#41
becuase of the dc coupling, i think i am going to have to use the 6dj8, realistically i havent got enough HT to play with.

The other option is to stick a cap in there, forget the dc coupling and then i have plenty of voltage to play with, and can use 6n1p's through out. question is, having 3 rc networs within the feedback loop... am i asking for trouble?
 
#42
having 3 rc networs within the feedback loop... am i asking for trouble?
Click to expand...

Yes.

Let's look at the possibilities:

1. Higher B+: the driver stage doesn't need a lot of current, so you conceivably could use a separate transformer to generate a high B+ at low current without a lot of cost/space penalty- this is what I did for the Red Light District amp. This also carries a stability advantage.

2. Lower plate voltage on input tube and diddling of operating points: with 70V on the first plate, the cathodes of the next stage will be at ~72V. Plates will be at ~220V. With 100k and 1.6mA current (still a good place to run them), your B+ requirement has dropped to 380V.
 
#46
I thnk i will need the gain of the 12at7.

just done through my valve collection (read junk box) and i have some nice brimar 12at7's (6061) and some jan phillips 6922s. so if the circuit looks reasonable, i need a mouser order for the resistors and caps and i am good to go. the rest is all from a donor amp.
 
#49
Well, its all wired up. Fired it up, and it makes music...... but

The CSS of the 12at7 isnt working correctly. I am only getting about 150v on the anodes.

The led biasing of the 6dj8 is fine, i get exactly 70v on the plate.

anyone have any suggestions for the css?
 
#50
adamus said:
Well, its all wired up. Fired it up, and it makes music...... but

The CSS of the 12at7 isnt working correctly. I am only getting about 150v on the anodes.

The led biasing of the 6dj8 is fine, i get exactly 70v on the plate.

anyone have any suggestions for the css?
Click to expand...

Check voltages to see if those are close to the design nominal values. Looks like it's pulling too much current for some reason. Could be a poofed transistor or the reference voltage is too high.
 
#51
Yes, Check the voltage at the base of your NPN on the CCS.

If your red LED is 2 Volts, you will get over 4 mA on the Css

2.1 mA * 100K = 210V drop across R5/6 Subtract that from B+ of the stage gives 390v - 210V = 180V at the plate.

I would look at CSS current and B+' first.
Doug

I see Miles beat me to it. 🙂

My college professor was old school and would have given me grief for a 3 significant digit voltage on a schematic. With 10% variation on the line voltage, specifying 1% voltages is little more than wishful thinking. 🙂
 
Last edited:
#52
checked the volatges in the CSS - look ok.

Base emitter voltage of the bc547 is 0.62v. The LED is dropping 1.82v, over 375 ohms that setting the current through the bottom resistor as 1.2/375 = 0.0032 amps which split between the two is 1.6 ma.

So i should be seeing 160v drop on the 100k resistors, but i am actually seeing nearly 250v+

both channels have the same issue, so its systematic. Both CSS are ok out of circuit..... anything wrong with the topology?
 
Last edited:
#53
Looking at it, the bias current for the LED is also coming from the cathodes.

Determine the additional current by calculating the current through the 25K resistor.

My guess is the first resistor would be to B+ and be much larger to limit the LED/bias current to a reasonable value.

Doug
 
Last edited:
#54
I have 55.6v over the 15k (change from schematic)

55.6/15k = 3.7ma through the 15k resistor

I think this means i need to use B+ instead of taking it straight from the cathode, otherwise i wont get low enough to keep the led happy - Think i am getting there...... bloody novices!
 
Last edited:
#57
and it works, all voltages are as designed (within 5v).

I'll try introducing some feedback, its slightly too loose at the moment, and the treble ever so slightly sibilant. I also dont have a css anode on the 6dj8, so i may try that too.
 
#59
Now i just need to get rid of the low level hum.

all taken as peak to peak.
Powersupply sims as 5v ripple . on the anode of the 6dj8 (first stage) there is 4mv of 50 hz. The next stage presents 0.75v of 100hz

speaker terminals present 30mv of 100hz which is slightly too loud.

Any ideas of where to start hum hunting?

Only change from schematic is that thee is no nfb at the moment
 
#60
SY said:
I will not be popular for saying this, but an output stage CCS is not generally a good idea, especially if you're going to clip the amp every once in a while (as we all do, at least all of us who don't have 102dB/1W speakers).
Click to expand...

Agreed.
If one is contemplating using this with fixed bias I've tried this with high gm 6550 and Kt88 types and although tempting, in most cases runs into instability; as the CCS offers a near infinity undamped AC impedance.
So stick with resistors.

richy
 
Status
Not open for further replies.
Top Bottom