6DJ8 biasing question

[jcc147]

Hi,

I'm new to tube and would like to pick this circuit as my first tube project.

Could some one help to understand the biasing method applied on 6DJ8?
Here is my own interpretation:
1. the 2SK117 (BG103) acts as a CCS and it's for active load purpose.
2. 100ohm(R102) is for 6DJ8 self biasing
My question is what current we can expect ? Is it determined by the trim pot R110? or 100ohm(R102)?

An externally hosted image should be here but it was not working when we last tested it.

Thanks!

 

[dsavitsk]

The trimmer sets the current. There is a bit of an explanation on how a plate load CCS works here.

 

[kenpeter]

My question is what current we can expect ?
Is it determined by the trim pot R110? or 100ohm(R102)?

Whichever bias resistor achieves equilibrium
at the lower current. Sets limit for the totem.

If roughly the same, whichever device has the
higher transconductance. In this case, the 2SK.

 

[jcc147]

The
I breadboard this circuit and set the pot to get 1mA.
After hooking CCS to the plate with the same pot value, the current has only ~0.15mA measured at the point where current going into 2SK117(S pin). Why is it not 1mA as I expect? And the current makes very less change when adjusting trim pot.
Is the 100ohm too small?

 

[jerluwoo]

The load on the ccs will also need to draw 1ma. The 6DJ8 is probably only drawing the .15ma you stated. Use a pot for the self bias and then adjust it to get 1/2 b+ at the plate, then measure the pot and use that value for the self bias resistor.

 

[Michael Koster]

The
I breadboard this circuit and set the pot to get 1mA.
After hooking CCS to the plate with the same pot value, the current has only ~0.15mA measured at the point where current going into 2SK117(S pin). Why is it not 1mA as I expect? And the current makes very less change when adjusting trim pot.
Is the 100ohm too small?

100R is an OK value for this low voltage op point.

At 1mA there should be 100mV across the 100R in the cathode and 8-10V
or so from cathode to anode.

How are you measuring the CCS current when you set the pot and then in-circuit with the 6dj8?

Michael

 

[jcc147]

How are you measuring the CCS current when you set the pot and then in-circuit with the 6dj8?

Michael

I set up a simple CCS circuit at the breadboard corner, then move the pot into the main circuit.

Is 100ohm not too small? got ~0.1-0.15mA on 6DJ8, and ~0.25mA on ECC86.

If 100ohm is ok, does that mean I should trim the pot to have ~0.1-0.15mA on CCS?

 

[SY]

Understand that this circuit will have rather high distortion and very poor overload. Low plate voltage, low bias, grid current...

 

[Michael Koster]

I set up a simple CCS circuit at the breadboard corner, then move the pot into the main circuit.

Is 100ohm not too small? got ~0.1-0.15mA on 6DJ8, and ~0.25mA on ECC86.

If 100ohm is ok, does that mean I should trim the pot to have ~0.1-0.15mA on CCS?

I was wondering how you are making the current measurement;
are you measuring the voltage across R102 and inferring the current,
or are you using an ammeter. Where are you connecting the probes
to measure the current?

From the 15V expected at the source of K310, I'd say it was
intended to operate the 6dj8 at a little over 2mA. this would give
a little over 200 mV across R102 and about 18V at the 6dj8 anode.

Michael

 

[jcc147]

I was wondering how you are making the current measurement;
are you measuring the voltage across R102 and inferring the current,
or are you using an ammeter. Where are you connecting the probes
to measure the current?

Michael

I used a ammeter and connect it between V+ and source of sk117.

I tried a BJT based CCS instead of 2SK117. Here is the circuit I referred:

An externally hosted image should be here but it was not working when we last tested it.

For some reason, I could get 1mA plate current after switching to BJT CCS method. But next issue comes up about the V plate.
V plate didn't change at all ( or very small) when adjust pot at cathode.

 

[jerluwoo]

I did some experiments with a circuit close to that using a 12au7. It sounded much better running the tube in A2 mode. Very difficult to get decent current or sound quality for that matter out of it at such low plate voltage. A low voltage tube would work much better. As stated above it will give some fairly high distortion since the tube will always be operating in the very nonlinear portion of the plate curves. A 1k pot in the cathode would be better than the 3.3k or 2.2k shown in the schematic. Based on the 6DJ8 datasheet plate curves you will need something like .6 to .7 volts to get 1 ma at around 15volts on the plate, that being if the tube still tests strong. A used tube may need even less to get any current out of it, which limits the amount of input and thus available gain before grid current even further. Using a high voltage b+ for the tube and then capacitor coupling to an output stage will no doubt give better results than the circuit above.