Tube DIY Asylum

Do It Yourself (DIY) paradise for tube and SET project builders.

Return to Tube DIY Asylum


Message Sort: Post Order or Asylum Reverse Threaded

Cathode bypass cap removed and I like it

50.80.146.114

Posted on January 10, 2017 at 15:12:10
banpuku
Audiophile

Posts: 957
Joined: January 19, 2006
Contributor
  Since:
March 30, 2017
I removed the cathode bypass cap on my 2a3 final tube. I like the sound, although I have to turn up my volume somewhat to compensate for the lower output. The sound is much cleaner and leading edges / transients are more lifelike, especially with piano.

Now, that said, should the cathode bypass resistor (1K ohms) be removed as well? I currently have it still in place from the filament supply transformer center tap to ground. Not sure if I should leave it in place or remove the resistor completely.

Thoughts?

 

Hide full thread outline!
    ...
RE: Cathode bypass cap removed and I like it, posted on January 10, 2017 at 15:20:49
Tre'
Industry Professional

Posts: 12276
Location: So. Cal.
Joined: February 9, 2002
If you remove the cathode resistor the tube will not have any bias voltage and it will burn up.


Removing the cathode bypass cap increases the plate resistance of the output tube and lowers the damping factor, a lot!

It is not a good idea.

There are other ways to apply feedback that won't ruin the damping factor of the amplifier.

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Cathode bypass cap removed and I like it, posted on January 10, 2017 at 15:35:48
Triode_Kingdom
Audiophile

Posts: 6147
Location: Texas
Joined: September 24, 2006
Removing the resistor will cause the amplifier to cease operating. There won't be a current path through the tube.



--------------------------
Buy Chinese. Bury freedom.

 

RE: Cathode bypass cap removed and I like it, posted on January 10, 2017 at 15:45:03
banpuku
Audiophile

Posts: 957
Joined: January 19, 2006
Contributor
  Since:
March 30, 2017
Thanks. Should the resistor value stay at 1K or be changed up/down?

 

RE: Cathode bypass cap removed and I like it, posted on January 10, 2017 at 15:51:33
Tre'
Industry Professional

Posts: 12276
Location: So. Cal.
Joined: February 9, 2002
Bias is DC.

The bypass capacitor prevents AC from forming across the cathode resistor when the tube is feed signal. This prevents the local current feedback that would otherwise occur.

The DC condition does not change when the bypass cap is removed so there is no reason to change the cathode resistor value.

Tre'






Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Cathode bypass cap removed and I like it, posted on January 10, 2017 at 17:02:15
rage
Audiophile

Posts: 511
Joined: December 17, 2010
what value was the capacitor you removed?

maybe replace it with a nice film cap of equal value.

 

RE: Cathode bypass cap removed and I like it, posted on January 10, 2017 at 17:17:41
megasat16
Audiophile

Posts: 207
Location: SoCal
Joined: April 15, 2015
Tre,

Are you saying the quoted RCA or any datasheets for the plate resistance is measured with the cathode bypass capacitor in place for A or AB circuits for the tubes such as 2A3? I meant the plate resistance is usually given in the data sheet and you can assume that plate resistance base on the operating parameter and use an appropriate cathode bias resistor and load the tube with a transformer with equal or a little higher primary Z.

Cathode bypass restores the degenerative feedback lost by the cathode bias resistor. AC flows through the resistor just as well as DC. The AC current is fed back into the GND so the feedback current is considerable reduced. The cathode bypass cap somewhat restores that lost current by acting like a buffer for the AC current.

My understanding is that you can safely remove the cathode bypass cap without effecting the plate resistance. The damping factor may be effected as it is calculated on the output from the amp and the amp without the cathode bypass cap has lesser (Volt or Power) output than the one with it. But this damping factor shouldn't matter much in the High Efficiency world when less than 1W of power is required for daily listening.
.
.
.Thou shall not stand where I type for I carry a bottle of Certified Audiophile Air and a Pure Silver Whip.

 

RE: Cathode bypass cap removed and I like it, posted on January 10, 2017 at 17:50:47
dave slagle
Manufacturer

Posts: 4789
Location: NYC
Joined: April 27, 2001
Well you have substantially changed the behavior of the circuit and I'll leave that banter to those seated firmly in their armchairs :-)

Since you like the results I think it would be interesting to look a little deeper and see if we can find a Why?

are you still using the ESL's?

dave

 

RE: Cathode bypass cap removed and I like it, posted on January 10, 2017 at 18:33:42
banpuku
Audiophile

Posts: 957
Joined: January 19, 2006
Contributor
  Since:
March 30, 2017
Yes, Dave, I am still using the Quad 63s. The reason why I like the sound better without the cap is because of the transparency and faster transients. The overall frequency sounds very even and rich. I listen 95% of the time to chamber music and solo violin/cello. So, richness and tonal purity is important.

Over the past couple of months I have tried the following cathode bypass capacitors:

50uF ASC Oiler
175uF ClarityCap TC
65uF ClarityCap TC
10uF Russian PIO K75
I even tried to bypass these caps with a 0.47 Russian PIO and Jupiter Condenser Copper Foil. The bypassing certainly helped (at least to my ears).

All of these caps imparted its own signature sound, depending upon the listeners one might like a given cap better than the other, but there was no real standout winner. Thus, why I removed the cap completely.

Also, the previous hum in my amp is nearly gone without the bypass cap. Another side benefit.

 

RE: Cathode bypass cap removed and I like it, posted on January 10, 2017 at 20:33:59
Tre'
Industry Professional

Posts: 12276
Location: So. Cal.
Joined: February 9, 2002
"Are you saying the quoted RCA or any datasheets for the plate resistance is measured with the cathode bypass capacitor in place..."

Yes.

"Cathode bypass restores the degenerative feedback lost by the cathode bias resistor."


If the cathode is grounded (no cathode resistor) then there is no degenerative feedback.

Once there is a resistor between the cathode and ground, if that resistor is not bypassed then the AC (along with the DC) would flow through that resistor.

When AC is flowing through the cathode resistor an AC voltage at the cathode is generated that is out of phase with the input signal at the grid.

The AC at the cathode bucks the input signal at the grid and is called local current feedback (degenerative feedback).

A properly sized cathode bypass cap insures that only the DC flows through the resistor. The AC flows through the cap because it's reactance represents a near dead short to AC and at the same time will not pass DC at all.

"My understanding is that you can safely remove the cathode bypass cap without effecting the plate resistance."

That is just simply not true. Current feedback increases the plate resistance. Voltage feedback decreases the plate resistance. Either way, the plate resistance is always effected by FB.

"The damping factor may be effected as it is calculated on the output from the amp and the amp without the cathode bypass cap has lesser (Volt or Power) output than the one with it."

No.

The damping factor is the reflected impedance of the primary divided by the plate resistance. It has nothing to do with voltage gain or power.

A amplifier that uses a 2a3 (assuming an operating point that yields a plate resistance of 800 ohms) connected to a output transformer that has a 2500 ohm primary has a damping factor of 3.125

If the output transformer were 5000 ohms the amp would have less power and less gain but the damping factor would be 6.25

Degenerative feedback will increase the plate resistance and that will decrease the (already marginal) damping factor.

Tre'


Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Cathode bypass cap removed and I like it, posted on January 10, 2017 at 22:32:14
Triode_Kingdom
Audiophile

Posts: 6147
Location: Texas
Joined: September 24, 2006
Don't change the value of the resistor unless the bias current is incorrect. You could try different types of resistors, though.


--------------------------
Buy Chinese. Bury freedom.

 

RE: Cathode bypass cap removed and I like it, posted on January 11, 2017 at 00:32:09
Alex M
Audiophile

Posts: 842
Location: Hampshire
Joined: February 9, 2001
Pat,

You don't give your reasons for removing the caps int he first place, but you reminded me that the late Allen Wright recommended in his Tube Preamp Cookbook that cathode bypass caps be removed for improved sound quality. The gross effect on gain will depend on the specific configuration and devices, of course.

Alex

 

Another solution - filament bias, posted on January 11, 2017 at 03:09:08
andy evans
Audiophile

Posts: 2973
Joined: October 20, 2000
This is more of a theoretical solution, since it requires something of a re-build. But it's the solution I found and have been happy with for a few years now. As said, cathode bypass caps affect the sound negatively. I tried all kinds of types when I used 2a3 outputs and just gave up - could not find anything I liked.

So instead of a 2a3 output, I used two DHT 4P1L trioded in parallel which is pretty much the same thing, other than that you get a welcome increase in gain from around 3 to around 11. Similar plate to cathode voltage of around 250v and OPT requirements.

But this gives you the opportunity to set up the tube properly with a low value cathode resistor of around 15 ohms. You need clean DC, so Rod Coleman regs and preferably a choke input DC supply to those. But it works perfectly - very clean and transparent.

It's a re-build because you need 2 more sockets and a heat sink for the regs, and also the cathode resistor needs to be a large wirewound type. But it's worth mentioning because it is a solution for this problem that works well. Filament bias won't work for a 2a3 or 300b, because the resistor would be too big and hot, but it suits the 4P1L perfectly.

 

RE: Cathode bypass cap removed and I like it, posted on January 11, 2017 at 06:52:41
dave slagle
Manufacturer

Posts: 4789
Location: NYC
Joined: April 27, 2001
OK.

I am not intimately familiar with the ESL63 but in general an ESL panel is a pure capacitance and the thing that makes them "difficult to drive" is actually a function of the step up transformer and the crossover that follows it. By removing the cathode bypass you have dramatically altered the behavior of the tube essentially increasing the RP from about 800 ohms to about 6K. This means your output impedance is now probably greater than the load it is driving and interestingly enough, some speakers actually like this situation.

The logical thing to do would be to use a mic and take frequency response measurement sweeps with and without the bypass cap and see if anything jumps out at you. My first guess would be that you would attenuate the HF info because of the higher output impedance working against a fixed capacitance. Unfortunately the 63's have a delay line in addition to a crossover so I fear simple theory goes out the window.

dave

 

The other obvious solution to cathode caps is fixed bias (nt), posted on January 11, 2017 at 07:21:52
Chip647
Audiophile

Posts: 1711
Location: The South
Joined: December 24, 2012

 

RE: Cathode bypass cap removed and I like it, posted on January 11, 2017 at 07:28:32
banpuku
Audiophile

Posts: 957
Joined: January 19, 2006
Contributor
  Since:
March 30, 2017
Tre, I am using the 3.5k Ohm tap on the OPTs. FYI.

 

RE: Cathode bypass cap removed and I like it, posted on January 11, 2017 at 08:14:43
Triode_Kingdom
Audiophile

Posts: 6147
Location: Texas
Joined: September 24, 2006
"The logical thing to do would be to use a mic and take frequency response measurement sweeps with and without the bypass cap"

That's really an excellent idea, and not too difficult if the primary concern is midrange and higher. Take the amp and one speaker to the backyard and have at it!



--------------------------
Buy Chinese. Bury freedom.

 

RE: Cathode bypass cap removed and I like it, posted on January 11, 2017 at 08:23:36
drlowmu
Manufacturer

Posts: 9706
Location: East of Kansas City
Joined: January 10, 2005
Resistor must stay, it self-biases the 2A3.

A single Rk bypass cap will never work on its own, it takes multiple caps, well chosen as to values and types, to bypass the Rk linearly and artistically well, so as to include ALL the musical content, in a pleasing way.

Its an art form, few know how to implement !! Have fun, I am too .


Jeff Medwin

 

RE: Cathode bypass cap removed and I like it, posted on January 11, 2017 at 08:57:43
banpuku
Audiophile

Posts: 957
Joined: January 19, 2006
Contributor
  Since:
March 30, 2017
Dave,

Thanks for the response. I don't have a mic to take frequency response info. Also, I have been using the 3.5K tap on my Hashimoto OPTs. The OPTs also have 2.5K tap, but after reading some of the replies to my post, it appears that I want to continue using the 3.5K tap, correct?

I am confused on fixed bias: by remove the cathode bypass cap, does this imply I have implemented fixed bias? Since I wont be tube rolling the 2A3s, should I consider fixed bias?

Pat

 

RE: Cathode bypass cap removed and I like it, posted on January 11, 2017 at 10:15:19
Tre'
Industry Professional

Posts: 12276
Location: So. Cal.
Joined: February 9, 2002
"Fixed bias" is a funny term.

Fixed bias is when you have a negative supply, that is usually adjustable, feeding the bottom of the grid resistor placing a negative voltage at the grid and the cathode is connected directly to ground (in your case the CT of the filament transformer would be connected directly to ground)

Self (or "auto bias", or cathode bias) is what you are doing (with or without a bypass cap).

The DC plate current flowing through the cathode resistor places a positive DC voltage on the cathode while the grid is at zero voltage DC.

The tube is now biased because the grid is negative WRT the cathode.

In the end the tube doesn't really care how the bias is accomplished as long as the grid is negative WRT the cathode by the correct amount.

Tre'


Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Cathode bypass cap removed and I like it, posted on January 11, 2017 at 11:30:36
banpuku
Audiophile

Posts: 957
Joined: January 19, 2006
Contributor
  Since:
March 30, 2017



Tre,

I found this schematic to help me understand fixed bias. What do you think of using a simple battery to set the negative bias voltage? I currently use a battery on my driver tubes, so I have no adversity to batteries. Unless you advise otherwise.

I am intrigued by the fixed bias method and want to hear what it sounds like. I just need to make sure I understand where the negative voltage should come from and how much negative voltage I should be using for the 2A3 tubes.

Pat

 

RE: Cathode bypass cap removed and I like it, posted on January 11, 2017 at 12:12:40
Tre'
Industry Professional

Posts: 12276
Location: So. Cal.
Joined: February 9, 2002
The amount of negative bias is dependent on how much plate to cathode voltage you will have and how much current you want flowing through the tube at idle.

The schematic left out the grid resistor. The grid resistor would go between the grid and the top of the battery.

If you change from cathode bias to fixed bias on your 2a3 there will be more plate to cathode voltage because the cathode will be at zero volts where as now it's at +45 volts or so.

That will change the amount of negative voltage you will need to reach the idle current that you want.

If you use a negative supply (instead of batteries) that delivers more negative voltage than you need then you can make it adjustable and you can set it where you need it.

I hope that made sense.

Tre'


Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Cathode bypass cap removed and I like it, posted on January 11, 2017 at 15:18:39
dave slagle
Manufacturer

Posts: 4789
Location: NYC
Joined: April 27, 2001
I'm not so sure how much it will help but you could do a frequency plot at the input of the speakers under both conditions and see if anything shows up. I have a hunch that the improvement you hear is due to the fact that you have drastically altered the behavior of the 2A3 and all of the suggestions of fixed / filament bias are attempting to solve a problem that may not exist.

dave

 

Page processed in 0.030 seconds.