• 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.

Cathode Bypass in a PP Output Stage

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Well, maybe this is an old topic, or maybe it's just too elementary, but I can not find a proper thread to follow. So here it is....

Learning from the great article by Mr. Lynn Olson here:
http://www.nutshellhifi.com/library/ETF.html

I have modded one of my amp from a "convetional PP"
ConvetionalPP.gif




to the "WE PP" style:

WEPP.gif


(I "borrow" some of the circuit diagrams from that site. And host these by another free site to avoid extra traffic on the original.)

I must confess that I can not figure out clearly the superiority in sound quality of the latter. Instant A-B comparison is impossible for me. I just think it's a better way to do the cathode bypass.

Now I'd like to move on.
How about this, even simpler differential type, with no cathode bypass:
DiffPP.gif


Before my rash actions, is there anyone who has already played with these? And how do they compare?



In addition, what's going on here:
WE92A300APPOutput.gif


What's the function of those "cross feeding" caps from the grid to the plate of the other side? Feed back?

Can I mod it a bit and use them on [cathodes]-[plates of the other side] with individual cathode resistors? Thus enlarging the current loop to the whole primary coil. Any benefit?
 
CLS said:
I must confess that I can not figure out clearly the superiority in sound quality of the latter. Instant A-B comparison is impossible for me. I just think it's a better way to do the cathode bypass.


I have done comparisons and was surprised at how much better the WE connection sounded. Only issue is that you need a higher voltage rating on the cap.

In the last circuit these are the "neutralising" caps. I think Thorsten has described the mechanism in detail somewhere here.

And have a look at this
 
#1

CLS said:
Well, maybe this is an old topic, ...

What's the function of those "cross feeding" caps from the grid to the plate of the other side?



Stop!
This is against the technical forum rules here. ;)

Technical forum rules:
µ = hfe_ hie = 1/hoe_
triodes don't change their characteristic at positiv grid voltages_
anode current is independent from g2 current in a pentode_
calculating the gain of a triode cathode follower you have to ignore the influence of µ_
Loftin White is cathode feedback_
[D]ignore the fourth circuit topology[/D]_
tubes should be replaced by mosfets_
neutralisation is forbidden_
continued

Read this thread: link

Kind regards,
Darius
 
Re: Re: Cathode Bypass in a PP Output Stage

analog_sa said:



I have done comparisons and was surprised at how much better the WE connection sounded. ....


It's good to hear that. I've been using the amp on and off these years mainly for bass (below 160Hz), and it's really difficult for me to identify the improvements after all the works on the 2 mono blocks. I lost the reference already. Maybe I should do it one channel at a time and use mono signal for comparing....

And thanks for the link, I'll be studying....
 
Administrator
Joined 2004
Paid Member
CLS said:


<snip>

In addition, what's going on here:
WE92A300APPOutput.gif


What's the function of those "cross feeding" caps from the grid to the plate of the other side? Feed back?

Can I mod it a bit and use them on [cathodes]-[plates of the other side] with individual cathode resistors? Thus enlarging the current loop to the whole primary coil. Any benefit?

I actually have a pair of pushpull 300B amplifiers I think would benefit from the Cmiller cancellation (VTV article amp I did about 10yrs ago) that the WE topology provides - I just need to find some caps with the requisite voltage ratings - if I ever do it I will report back here. The primary benefit should be much more extended HF response as the driver stage no longer has to drive all of the output tube miller capacitance. IMO It is safer to err a bit on the low side cap value wise in order to avoid the possibility of creating an HF multi-vibrator. (It is positive feedback afterall..)
 
I experimented with the WE connection in the past (actually I build both the Amity and Karna) and found that the connection does not work when implemented like it is in those amps. I could neither hear, nor measure a difference in harmonic distortion.

It can work however, it's purpose being to reduce 3rd harmonic which is very benificial in a PP amp. If my memory serves me correctly, then for it to work you need to split the cathode resistor into two series resistors (or a pot) and connect the B+ to cathode cap at that junction (wiper). The values need to be tuned but are not very critical.
 
I'm in the middle of testing some of these options on my 300b PP amplifier also. Here's how I understand these circuits to work.

I think the diagrams you give are a bit simplistic to explain all of the different imbalances in signal, gain, signal swing + to -, and DC that enter into the picture, particularly as relate to cancellation, reinforcement, or conversion of distortion products. It's particularly important to remember the triode load-line asymmetry in this analysis. This is a factor even if one were to perfectly match the signal path from side to side.

The WE connection cancels third harmonic distortion by feeding the AC distortion residual from the B+ (taken from the OPT CT) back to the cathode. So I think it is a form of cathode feedback.

Basically, when an amplifier is driven to higher signal levels and distortion increases, the waveform peaks draw more current than the average, supposedly balanced, B+ current. These current peaks, through the finite impedance of the power supply, cause an image of the distortion waveform to appear at the OPT center tap.

This distortion current, when returned to the common cathode connection, becomes a common mode negative feedback that tends to restore the flattened peaks of the voltage waveform (f3 distortion) which occur in the circuit due to the current peaks at large signal.

You may only notice the difference at higher signal levels or on "musical transients" which are much louder than the average in order to drive the amplifier into third harmonic distortion. Impedances need to be balanced for it to work well at all.

On the other hand, the differential output stage almost guaranteed third harmonic distortion at large signal levels, yet it sounds great also.

The other circuit consisting of cross-coupled caps, anode to opposite grid, is to cancel the miller effect. Basically, a second capacitance, equal to the tubes anode-grid capacitance, is inserted to the anode of the opposite tube. I think you need to factor (mu-1/mu) in there also for the grid swing, but what you end up with is the two anodes pulling opposite directions with the same effective capacitance, resultant = 0 net reactance. Also known as neutralization.

Cheers,

Michael

PS hope I didn't just step in something...
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.