• 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.

Output transformer in the cathode?

Status
Not open for further replies.
Jump to Latest
#3
I have built a few cathode follower amps. There are two obstacles to overcome.

The output stage has no voltage gain so you need a driver that can swing several hundred volts peak to peak. This isn't as bad as it sounds, it just means that your driver is a mini output stage. I have used a choke loaded 6EM7, a choke loaded 6SN7 and a CCS loaded 45 with a 500 volt supply.

The cathode will swing the entire output voltage range, again several hundred volts peak to peak. This will violate the heater to cathode voltage rating of most mere mortal tubes.....violate it bad enough to make sparks inside the tube. The 6528 and 6336A were the only tubes that I know of that may survive. I have never tried the 6C33. The obvious solution is to have a heater winding for each tube and wire it directly to the cathode, but this mandates one output tube for each channel, and a separate winding for the drivers.
 
#4
I forgot the 6c33c doesn't have a dual cathode..🙄
(Just dual heaters)..so it would have to be another tube type..😀

Yes the Heater cathode voltage..I had overlooked quite a few things..(Not so well planned me thinks)
I will have to research more, so it can be done in some cases.


Regards
M. Gregg
 
Last edited:
#5
#6
M Gregg said:
Cathode-Follower Power
Click to expand...

Could you use a triode in this mode?

Yes.
Are there any problems with the driver?

Yes. The A Number One problem here is that a cathode follower has less than unity voltage gain. Your driver needs to swing the whole output. What you gain in terms of linearity of the final, you might lose in distortion in the up front stages. An added complication is possibly having to use two power supplies: a high voltage supply for the front end, and a separate, lower voltage PS for the finals. That's all but guaranteed if you're using the series pass triodes (6AS7, 6C33) that like lower VPK's, pseudotriodes made from HD finals (limited V2K specs).

It's the same problem you have if using low-u power triodes, or screen drive with HD pents. An added complication, but not necessarily fatal to the design.

You also have to watch out for that VHK spec, as you would any time you're running the cathode far from the ground potential. You'll either need to float the heater(s) or provide protective heater bias.

I'm thinking SE (Has anyone tried it?)

Never did.
 
#7
Why build a cathode follower output stage in the first place? Are there some really great sonic advantages? Seems like a ton of work and maybe no better sonics over a regular plate loaded output stage, especially if output transformers are involved in both cases?

On the other hand, cathode feedback like McIntosh did, I get.
 
#9
kward said:
Why build a cathode follower output stage in the first place? Are there some really great sonic advantages? Seems like a ton of work and maybe no better sonics over a regular plate loaded output stage, especially if output transformers are involved in both cases?

On the other hand, cathode feedback like McIntosh did, I get.
Click to expand...

I have a cathode follower amp using triode connected SE PL504 . Rock solid bass without resulting to loop feedback is one advantage . Another is that during complex passages the sound does not 'fall apart' like with a normal SE amp . I know 300B etc SE amps do simple jazz or 'little girl with a guitar' type music well , but these amps don't excel at everything while cathode follower amps seem to have no trouble with all types of music . Another plus is the output transformers , these do not need to be pricey units to get good results . CF amps just need a half-decent driver which incidentally a SE power amp can be used cap-coupled from the anode of the output valve

316a
 
#10
kward said:
Why build a cathode follower output stage in the first place? Are there some really great sonic advantages? Seems like a ton of work and maybe no better sonics over a regular plate loaded output stage, especially if output transformers are involved in both cases?

On the other hand, cathode feedback like McIntosh did, I get.
Click to expand...

With a cathode follower output, you can use a transformer that has less of a turns ratio, resulting in greater bandwidth and lower output impedance. Also, be aware that the DC resistance of the primary may not be enough to bias the tube, so you may also need a grid bias supply.

McIntosh used combined plate load and cathode feedback in some of their push-pull output stages. Sowter makes some transformers that could be used for that purpose.
 
#12
Yeah, maybe not, since the cathode needs a high load. I suppose what I was thinking of is that the transformer is driven by a much lower source impedance, which then results in greater bandwidth, lower distortion, and lower output impedance.

Still, the drawbacks remain as described above.
 
#13
Why build a cathode follower output stage in the first place? Are there some really great sonic advantages?
Click to expand...

Like lower output Z? Or inherent 100% degenerative feedback? Yes.

Some could say that there were enough advantages to ditch tubes practically completely in favour of semiconductor circuits that lend themselves much better to follower output -type power amplifiers.
 
#16
dirkwright said:
With a cathode follower output, you can use a transformer that has less of a turns ratio, resulting in greater bandwidth and lower output impedance.
Click to expand...

That's a very common misconception. If you want to realize the distortion and impedance advantages of a cathode follower, you have to load it pretty much the same way you would if you were putting the load in the plate.

There's a very nice, clear discussion of this in Crowhurst's excellent "Understanding Hifi Circuits." I think that Morgan Jones touches on it in VA3/VA4 as well. And of course, it's done in detail in RDH4.
 
#18
dirkwright said:
With a cathode follower output, you can use a transformer that has less of a turns ratio, resulting in greater bandwidth and lower output impedance.
Click to expand...

Actually, you can't do this. The loadline remains the same, regardless of how you connect the load. The cathode follower can not, and does not, make any VT into something it was never meant to be: a high current, low voltage device. It can present a Lo-Z source, but only to a Hi-Z load.

The lower source impedance can help with low frequency performance, as it reduces the magnitude of the resistance connected to the primary of the OPT, making XLp a greater percentage of primary load to both reduce the low frequency cutoff, and making a dynamic loadline less like a circle to keep the plate out of the nonlinear/cutoff region.

It doesn't change the actual loadline itself.
 
#19
SY said:
That's a very common misconception. If you want to realize the distortion and impedance advantages of a cathode follower, you have to load it pretty much the same way you would if you were putting the load in the plate.

There's a very nice, clear discussion of this in Crowhurst's excellent "Understanding Hifi Circuits." I think that Morgan Jones touches on it in VA3/VA4 as well. And of course, it's done in detail in RDH4.
Click to expand...

Um, see post #12.
 
#20
Miles Prower said:
Actually, you can't do this. The loadline remains the same, regardless of how you connect the load. The cathode follower can not, and does not, make any VT into something it was never meant to be: a high current, low voltage device. It can present a Lo-Z source, but only to a Hi-Z load.

The lower source impedance can help with low frequency performance, as it reduces the magnitude of the resistance connected to the primary of the OPT, making XLp a greater percentage of primary load to both reduce the low frequency cutoff, and making a dynamic loadline less like a circle to keep the plate out of the nonlinear/cutoff region.

It doesn't change the actual loadline itself.
Click to expand...

and, yet again, if you bothered to read more, you'd see that I corrected this in post #12.
 
Status
Not open for further replies.
Top Bottom