Ideal Grounding...Question for MerlinB

[jrfrond]

I know that a lot of old designs use can caps. I've repaired thousands of them. Despite the fact that they are again available, I would never use one in a new build.

Someone said they wanted to "get rid of the preamp filter cap all together". You can't. The purpose of that cap is MAINLY to decouple the stages, i.e. remove audio signals that get superimposed on the B+ line of a tube stage to keep them from leaking to other stages, causing unintentional feedback paths.

As far as grounding is concerned: I spent many sleepless nights debugging the grounding scheme in my main amp design that LOOKED technically perfect. Star-ground systems are not a panacea for grounding issues. Look at you typical BF amp, which are usually dead quiet. It is far from star grounding, but I am sure that Fender engineers worked hard on it.

For me, the key points with grounding are to keep the heater supply CT away from the preamp grounds, and essentially divide the grounding scheme into low-current/high-current. I view everything as celectron flow, which states that all current emanates from the LOW side of the circuit, ground in this case. If you were to ground your heaters to the same node as the 1st stage preamp ground, the heavy AC current draw of the heaters would enter the stage through the cathode and be amplified right along with the audio, and you would never see it entering the stage at the grid.

To me, Fender got it TOTALLY right with the BF amps, to wit, separate ALL stage grounds to avoid accidental coupling through the cathodes.

[Wilder Amplification]

Someone said they wanted to "get rid of the preamp filter cap all together". You can't.

That was meant tongue-in-cheek...should've put the smileys in to note that.

For me, the key points with grounding are to keep the heater supply CT away from the preamp grounds

I run elevated heaters on my builds now...elevated via a +40VDC or so reference. But I get what you're saying....keep all the high current grounds clear over on the opposite end of the chassis from preamp grounds

This last build I did just that. It's a hot rodded 2203 circuit. I even lifted the ground buss that would normally be on the back of the pots off the chassis, grounded the entire preamp to that, then ran 1 wire from that buss to the preamp filter cap negative, which then couples to the chassis via a ground lug at the filter cap mounting bracket, which takes preamp current completely off the chassis (except the negative power supply feed of course). It's dead quiet...at the volume where you start to hear ANY power supply hum you're ripping peoples' heads off.

[Gaz]

I've recently started using radials mounted like multi-section cap cans to save space on the turret board. Works well, and makes the amp look more 'complicated' from the the top side Not to mention, high quality computer grade caps can be found real cheap on Ebay. Unless you guys would rather pay $14.00 for a 50/50 F&T!

[Merlinb]

If you were to ground your heaters to the same node as the 1st stage preamp ground, the heavy AC current draw of the heaters would enter the stage through the cathode and be amplified right along with the audio, and you would never see it entering the stage at the grid.

Hmm, well whichever kind of current you prefer to use, none of it flows in the heater CT!

[trem]

For me, the key points with grounding are to keep the heater supply CT away from the preamp grounds, and essentially divide the grounding scheme into low-current/high-current. I view everything as celectron flow, which states that all current emanates from the LOW side of the circuit, ground in this case. If you were to ground your heaters to the same node as the 1st stage preamp ground, the heavy AC current draw of the heaters would enter the stage through the cathode and be amplified right along with the audio, and you would never see it entering the stage at the grid.

I never even considered this (I am not a tech). So I imagine the same applies to the "Virtual" center tap using Two 100 ohm resistors.? Assuming the PT supplied 5 amps to the heaters, all this current flows through the center tap.?
Thank You

[raiken]

Assuming the PT supplied 5 amps to the heaters, all this current flows through the center tap.?

No, no current flows through the center tap, it all flows to the filaments and back, in a loop, none exits the CT because there is no return path (unless you want to include capacitive/inductive coupling or other high-frequency effects). The CT is only there to provide a reference voltage for the filament string.

Randall Aiken

[Wilder Amplification]

No, no current flows through the center tap, it all flows to the filaments and back, in a loop, none exits the CT because there is no return path (unless you want to include capacitive/inductive coupling or other high-frequency effects). The CT is only there to provide a reference voltage for the filament string.

Randall Aiken

Yep...and lifting it +40 above ground is handy for reducing heater - cathode voltage (Vk/f) in cathode follower circuits as well as PI circuits, which allows you to run higher preamp voltages without exceeding that rating.

[anson]

For the reduction of hum, I've found it worthwhile to twist the current supply lead with the current return lead for a given circuit, so as to minimize any "loops" that can act as antannae to xformer hum and stray interferance. For example, take the HT lead that runs off the power suply filter caps to the OT. That lead supplies voltage to the plates of the power tubes, which draw electrons form the power tube cathodes. So I twist that HT lead with the power tube plate and cathode leads and have that cathode lead return to the ground side of the filter cap from which the HT lead came. There are other similar sceneros all over an amp, in the power supply and in the preamp that can benefit from the same treatment.

[trem]

So what is the "best" way for me to approach this V4-B.? I replaced the E-Caps and tried to clean up some other "mods". The vertical terminal strip next to the Series/Parallel can cap connection was put in by me. I can take it out if you guys think it is problematic. But there was a lot of connections to be made in that area, and this made things easier for me.
The hum balance pot was bad. I have never used those on an amp before, so I went with the 2x100 ohm resistor heater CT. I have them hanging off the first power tube, and terminated at the mains ground.
Thank You
(would it be better if I started my own thread for this....I thought it tied in to the former. sorry about that)

http://i111.photobucket.com/albums/n...4-Bcaps001.jpg
http://i111.photobucket.com/albums/n...4-Bcaps002.jpg
http://i111.photobucket.com/albums/n...4-Bcaps003.jpg
http://i111.photobucket.com/albums/n...4-Bcaps004.jpg
http://i111.photobucket.com/albums/n...4-Bcaps005.jpg
http://i111.photobucket.com/albums/n...4-Bcaps006.jpg
http://i111.photobucket.com/albums/n...-Bcaps001A.jpg
http://i111.photobucket.com/albums/n...-Bcaps002A.jpg
http://i111.photobucket.com/albums/n...-Bcaps003A.jpg

[Wilder Amplification]

So what is the "best" way for me to approach this V4-B.? I replaced the E-Caps and tried to clean up some other "mods". The vertical terminal strip next to the Series/Parallel can cap connection was put in by me. I can take it out if you guys think it is problematic. But there was a lot of connections to be made in that area, and this made things easier for me.
The hum balance pot was bad. I have never used those on an amp before, so I went with the 2x100 ohm resistor heater CT. I have them hanging off the first power tube, and terminated at the mains ground.
Thank You
(would it be better if I started my own thread for this....I thought it tied in to the former. sorry about that)

http://i111.photobucket.com/albums/n...4-Bcaps001.jpg
http://i111.photobucket.com/albums/n...4-Bcaps002.jpg
http://i111.photobucket.com/albums/n...4-Bcaps003.jpg
http://i111.photobucket.com/albums/n...4-Bcaps004.jpg
http://i111.photobucket.com/albums/n...4-Bcaps005.jpg
http://i111.photobucket.com/albums/n...4-Bcaps006.jpg
http://i111.photobucket.com/albums/n...-Bcaps001A.jpg
http://i111.photobucket.com/albums/n...-Bcaps002A.jpg
http://i111.photobucket.com/albums/n...-Bcaps003A.jpg

Where you grounded the heater CT is fine. No current flows in the CT (or the false CT in your case with the 100R resistors)...it's just there to give a zero volt reference. The heaters themselves are "floating" so they don't see the ground reference voltage which is why no current flows in the CT, but the zero volt reference at the CT places the heater/cathode voltage (Vk/f) on the power tubes at 1/2 the heater voltage assuming grounded cathodes on the power tubes. The Vk/f for the preamp tubes would be the difference between 1/2 the heater voltage and the cathode voltage unless it's a DC coupled cathode follower stage with the plate tied direct to the rail, in which case it would be much higher.

In the case of a standard common cathode triode gain stage, the heaters would still be more positive than the cathode by a volt or two since cathode voltage is usually around 1-2 volts above ground. However, in a long tailed pair phase inverter stage the cathode is more positive than the heater since the cathode is usually referenced around 35-40 volts due to the higher value tail resistor (usually 10K (Marshall) or 22K (Fender)) and a DC coupled cathode follower stage would put the cathode much higher than that. This is where DC elevation of the heater supply comes in handy when you want to run higher preamp node voltages without exceeding the Vk/f rating of the preamp tube used in the cathode follower stage.

[Joey Voltage]

(usually 10K (Marshall) or 22K (Fender))

It's about 15K for marshall, you have to take into account the 4.7K "bottom tail". It's not just part of the -Fb circuit, it is providing a dual purpose as there is a drop that occurs across both of these. You can ignore it for most fenders because it is very small in comparison to the very large 22K.

To the OP, It may be best to start a new post. It will probably get more attention.

[Wilder Amplification]

It's about 15K for marshall, you have to take into account the 4.7K "bottom tail". It's not just part of the -Fb circuit, it is providing a dual purpose as there is a drop that occurs across both of these. You can ignore it for most fenders because it is very small in comparison to the very large 22K.

To the OP, It may be best to start a new post. It will probably get more attention.

Well yes...I was thinking of the tail resistor value by itself. The main point being is that it sets the cathode voltage up a bit higher than a standard common cathode triode stage.

Although the cathode voltage to ground is a bit higher, the nature of the bootstrap network elevates the grid reference in such a way as to place the grid 1 volt below cathode voltage so the stage biases normally just over 1mA, similar to how the typical Rp = 100K/Rk = 820R stage biases up (470R is used due to having two stages pull current through it so the bias ends up being roughly the same as a single stage with a 940R used as Rk).

[Joey Voltage]

Well yes...I was thinking of the tail resistor value by itself.

the tail is comprised of the added value of the 10K+4.7K which is an increase of about 20%. Now does this change the heart of the matter that the Cathode is sitting at a high potential in either case, no, although I don't even know what this truly had to do with the subject....I don't think the principal of DC elevation, and grounding are very convoluted. I was mainly pointing it out to you because many people do not consider this.

Although the cathode voltage to ground is a bit higher, the nature of the bootstrap network elevates the grid reference in such a way as to place the grid 1 volt below cathode voltage so the stage biases normally just over 1mA, similar to how the typical Rp = 100K/Rk = 820R stage biases up (470R is used due to having two stages pull current through it so the bias ends up being roughly the same as a single stage with a 940R used as Rk).

now we are way OT. of course it does, and there are many different ways you can visualize this. whether it is bootstrapped or not is beside the point, The relationship between the grid potential relative to the cathode is exactly the same as it is for any CC/fixed bias gainstage.