Fixed bias, cathode feedback?

[mctavish]

I have a SE KT88/6550 amp loosely based on the tubelab simple amp. It uses a CCS loaded 12AT7 driving a fixed-bias KT88 tube @ 450v and around 60ma.

I have it built with a switch for triode/UL mode because with fixed bias I can swap in all sorts of tubes. ( I prefer the EL34 in triode; the 6550 in UL.)

A Question, I would like to try cathode feedback with the 6550 in UL. I have Hammond 1628SEA transformers with 0-4-8-16 ohm secondaries.

I would attach the 16 ohm secondary to the cathode of the output tube so that it is grounded through the common.

Will the standing current harm the secondary?

I this a bad idea because it will impose 60ma. current on the speaker voice coil?

I know tubelab cap couples the cathode feedback but that is in cathode bias.


I cant seem to wrap my head around the particulars...

Any guidance?

 

[Miles Prower]

Cathode lNFB is usually used with PP finals, so that there isn't that DC magnetization problem. With a SE OPT, you might try connecting it so the DC fields of the primary and secondary offset each other. Maybe that prevents core magnetization, or maybe not. YMMV.

 

[mctavish]

Thanks Miles. That was my thought as well. A bit of experimentation will be called for



Will the standing DC current be a problem for the speaker? this is the one that I am having trouble conceptualizing.....

 

[Hilmar53]

60mA DC through the Voicecoil is not so good i think.
But when you have separate sec. coils in your OT then you can perhaps take one coil for the feedback and the rest for the speaker.
Maybe you show us a little schematic how your OT looks.
Since you are from Felix Austria:
Ich glaub nicht, daß 60mA Gleichstrom gut für den LS ist, wenn Du aber,wie zum Beispiel bei Lundahl,separate secundär Wicklungen hast,könnte es gehen.
Best wishes,
Hilmar

 

[mctavish]

I have the Hammond 1628SEA outputs. I appears to be one secondary winding that is tapped.

 

[SY]

What's the DCR of the secondary? You might want to calculate how the current divides between transformer and voice coil...

 

[Ketje]

First,why fixed bias.Has only drawbacks for SE ,exept somewhat more Vb.
The 60mA in the secundary is almost completly shortcircuited by it's DC resistance.
Certanly not try to cancel the magnetic field with the current from the cathode !
If you do that the result is positif feedback and oscillation.
Don't forget,after that mod the input sensitivity of the amp will be smaller if there was no overall feedback.
Mona

 

[Tom Bavis]

For example, if the 8 Ohm winding is 0.250 Ohms DCR and the 8 Ohm speaker is 5 Ohms DCR, the speaker will get about 3 mA, a 15 mV offset.

The DC current has only a small effect on core magnetization, since there aren't many turns. Cathode feedback effectively increases the primary impedance. New Z = (sqrt(Zin)+sqrt(Zout)^2

 

[jazbo8]

I think it could work with some fiddling... an old post from tubelab:
"If the transformer has multiple impedance taps, you hook the speaker up where it belongs, and try each tap for CFB. Pick the one that works the best in your situation. If it happens to be the same one that the speaker is on, use it."

From here.

 

[Ketje]

For example, if the 8 Ohm winding is 0.250 Ohms DCR and the 8 Ohm speaker is 5 Ohms DCR, the speaker will get about 3 mA, a 15 mV offset.

The DC current has only a small effect on core magnetization, since there aren't many turns. Cathode feedback effectively increases the primary impedance. New Z = (sqrt(Zin)+sqrt(Zout)^2

Indeed,looking from the tube you have anode-ACgrd-DCgnd-cahode.The first two is the primary and the later the speaker winding.Effectively puting the two in series.
Little bonus,if you don't have UL,now there is (somewhat)The interconnection ACgnd-DCgrd is also the ACgrd of the screengrid,puting it on a tap of the OPT.
Mona

 

[mctavish]

When I get home I will measure the DCR of the secondary. My goal here is a bit better damping factor in UL mode.

The speakers have a DCR of 7.6 ohms so if the secondaries are only a fraction of this than the miniscule dc on the voice coil should be negligible yes?

Ketje, I went with fixed bias simply for the ability to use different output tubes. With rectifier changes and bias changes I can use 6550. EL34, 6L6, ect.

This is not so easy with cathode bias.

 

[Ketje]

For experiments an adjustable bias is a good idea
With ageing you have to adjust the bias for time to time.Autobias well ..that's automatic
Mona

 

[costis_n]

The way to do it is ac-coupled.Ground one end of the secondary and couple the feednack to the cathode through a ~100μ capacitor. Positive of the capacitor goes to the cathode.

 

[mctavish]

The way to do it is ac-coupled.Ground one end of the secondary and couple the feednack to the cathode through a ~100μ capacitor. Positive of the capacitor goes to the cathode.

I dont see how this would work as I am using fixed-bias and the cathode is grounded.

 

[mctavish]

So I measured the dcr of the secondary; 1.1 ohm from common to 16 ohm tap. In other words, the standing current through the speaker vc will be really low.

So 60 milliamp across the 1.1 ohms would be only 0.066 volts.

I am going to try: cathode= > 10 ohm resistor=> 16 ohm tap=> common tap grounded.

The 10 ohm resistor is for measuring the cathode current. My speaker is on the 8 ohm tap.

I will report back with what I find.

 

[jazbo8]

Based on Tubelab's post from the above link - using the secondary tap for cathode feedback, hope I got it right...

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

 

[costis_n]

Based on Tubelab's post from the above link - using the secondary tap for cathode feedback, hope I got it right...

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

Yes, that's it. If feedback turns positive though, you have to reverse it

Sent from my Jyayu using Tapatalk