Does it matter which cathode cap gets grounded to the ecap first?

[Mike_J]

In the process of connecting the ground side of the power tube cathodes to the negative end of the first stage ecap. Seems like a silly question but if I am not 100% sure of something I am going to ask the question. The way I intend to wire this is from the ecap to V5-cathode ground on to V4-cathode ground and then on to the PT bolt. Would anyone out there please confirm that it is okay to do this? I can't imagine there would be a problem with it but just want to ask those who are more knowledgeable than myself for piece of mind sake. Or is it peace of mind sake. Don't remember which but anyway getting off the subject.


Thanks
Mike

[Mike_J]

Had another thought. Might it be better to send the ecap ground and the power tube cathode grounds to the PT bolt separately?

[Mike_J]

While we are at it Aiken says, "If the amplifier uses global negative feedback, there will have to be a ground path from the secondary common back to the phase inverter tail ground (or wherever the feedback is returned to).  This should be done via a wire from the ground of the output jack back to the ground of the stage the feedback is applied to.  The idea is for the feedback amp to amplify the difference between the feedback "hot" lead and the feedback ground lead, but nothing else." 


Let me start by saying I am confused by this. Not entirely certain what he means about global negative feedback but will look that up. If what he is saying applies to the 5f6a it sounds like the presence pot ground should be attached to the OT secondary ground on its way to the PT bolt. Does anyone else read it that way? If so, the attachment of the ground for the pot would need to be isolated from the other preamp grounds. Going to do a little more reading on global negative feed back.


Thanks
Mike

[pdf64]

Yes, the LTP is part of the power amp system.
A power transformer mounting faster is a poor choice for 0V to chassis connection lugs.

[Mike_J]

Yes, the LTP is part of the power amp system.
A power transformer mounting faster is a poor choice for 0V to chassis connection lugs.

Thanks pdf64. The 5k potentiometer is the ground end (second part of the tail resistor) of the LTP as I read it. One side of a 27K resistor goes to the same 5K pot. The other side of the 27K resistor goes to the tip of the output jack which is also connected to the positive side of the OT secondary. So should I remove the presence pot from the preamp grounds and run this ground to a ground position on the power end of the chassis. That is my interpretation by extension of what Aiken is saying because he is adamant that the OT secondaries be isolated from the preamp because of the current it carries.
Unfortunately that is only half of my confusion. Also sounds like Aiken is suggesting by saying "This should be done via a wire from the ground of the output jack back to the ground of the stage the feedback is applied to.  The idea is for the feedback amp to amplify the difference between the feedback "hot" lead and the feedback ground lead, but nothing else."
Where I am still confused is I have a separate 22uF @ 500VDC ecap for the LTP V3. From my interpretation of the LTP the presence pot ccw tab is the grounding point for the LTP. If you look at the 5f6a schematic the 10K tail resistor is combined with the 5K presence pot to form a 15K tail resistor which as I mentioned grounds at the 5K presence pot. Doesn't sound to me like Aiken would want the preamp grounded to it because as you mention the LTP is part of the power amp system. May be making a mountain out of a mole hill but I would like to understand this better.


Hot Blues Plate made a comment something to the effect that a circuit is a circle. Is it one circle or a number of them. If it is a number of them it seems like the trick would be to make sure everything that goes out positively to a circuit comes back together on the negative or ground side. I am not an EE, just a retired CPA but I am capable of learning once things make sense to me. Right now it is still a little fuzzy. Don't know where the OT ground should run to make the circle HBP was talking about.


Thanks
Mike

[Mike_J]

Have thought about where I am in this grounding exercise. V1 and V2 have four cathode resistors that need to be grounded to the filter cap that filters them. (I know it is an over simplification in that all the ecaps before it did filtering as well.)


Stage one's only question is do you tie the power tube cathode grounds to the ground side of the stage one ecaps. Other choice is to send the cathode grounds to the PT bolt by themselves and the stage one ecap ground individually to ground as well. Aiken is picky about the OT secondaries and the power tube cathode grounds. I think he mentioned it is because of the current involved in those areas, especially the OT secondaries. Have to keep those separated from the preamp area he is very clear about.


Stage two is very simple. It would normally go straight to the PT bolt or other power side grounding point if the PT bolt is not used. Because I have a potential voltage divider that would provide DC under the AC for the heater supply that is filtered by stage two I sent the ground to the turret that would be used as the ground point for the voltage divider and then off to the PT bolt.


Now the one I am completely confused about V3 the LTP. As it stands now the presence pot which is allegedly the end of the tail resistor for the LTP started by the 10K resistor in the LTP is tied to the grounds for the preamp, V1 and V2. I am sure it has worked well for a long time but I am doing this exercise from a theoretical as well as a practical perspective and from a theoretical perspective I am not sure it is the best option all though it may be. This is just an exercise after all.


As pdf64 said "the LTP is part of the power amp section." Aiken says they shouldn't be mixed. So this provides me with confusion which always gets me to ask questions.

The source of my confusion is as follows. As mentioned the presence pot. Aiken said something about tying the presence pot ground to the output jack ground if I understand him correctly which is an issue all of its own. This would mean the PI and output jack would be grounded together. Don't know that is at all what Aiken meant. Then there are some grounds sitting out all by their lonesome. They are the grounds for the bias circuit. Not sure if they should be tied to the LTP since they feed negative voltage to the junction of the 220K resistors in the LTP which goes to the grids of the power tubes. They come with their own smoothing caps I think which may answer the question and they should be off to the PT bolt by themselves. The .1uF coupling caps coming out of the LTP also impact the grids of the power tubes so there is a question in my mind as to where do these parts belong.

If anyone can help clear up my confusion please do.

Thanks
Mike

[Mike_J]

Okay, looking at the OT. There probably are only two choices for the grounding of the OT. First the OT primary CT is supplied by stage one so the first ecap ground is certainly a possible grounding point for the OT secondary ground. The other as mentioned is sending the OT secondary ground to the ccw lug on the presence pot and sending it straight to the PT ground lug which is also the LTP ground. Guess the third option is to combine both by sending the first stage ecap ground and the OT secondary ground and LTP ground which is at the presence pot ccw tab to exactly the same place, I mean the same hole wires touching at the PT bolt. Think this is the answer worth trying although I am not entirely sure why other than when you look at the 5f6a schematic it seems logical.                                                         

[pdf64]

Apologies but those are substantial chunks of text to work through, and it would be helpful if you could summarise the key points you need help with.

The principle of the star point is that all 0V returns in that system connect back only to there. The only exception being with 2phase rectification; there the winding’s CT should connect directly to the reservoir cap terminal.

It seems somewhat perverse to seek the ideal LTP balance and 0V arrangement yet determinedly insist on a rather poor choice for that star point’s chassis connection, one that will inevitably loosen over time.

[Mike_J]

Apologies but those are substantial chunks of text to work through, and it would be helpful if you could summarise the key points you need help with.

The principle of the star point is that all 0V returns in that system connect back only to there. The only exception being with 2phase rectification; there the winding’s CT should connect directly to the reservoir cap terminal.

It seems somewhat perverse to seek the ideal LTP balance and 0V arrangement yet determinedly insist on a rather poor choice for that star point’s chassis connection, one that will inevitably loosen over time.

The connection point will be somewhere other than the PT bolt but not sure where. Any suggestion would be greatly appreciated. Read somewhere that the PT bolt was a good grounding point. That is why I still muse the term ground at the PT bolt generically because I don’t know a better place for it.

The first time you told me the PT bolt wasn’t good I believed you. My focus has been on appropriate ground arrangements. Will switch to the best ground point and method of attachment about one second after I am satisfied with the grounding arrangement.


Thanks
Mike

[acheld]

I just choose a convenient point close to the PT and install a terminal strip.

I was just looking at http://sluckeyamps.com/misc/AB_Lite_II.pdf and you can see the grounding scheme as plain as day.

Very simple, and right to the point.

[Bieworm]

My grounding is simple and effective:

1. grounding point dedicated to ONLY the power cord from the wall. As near as possible to the entrance point of the wall power into the chassis. The connecting wire should be longer than the L an N wires in case you rip that wire out of the chassis so ground wire is the last one to tear loose.

2. Power stage ground terminal. A star ground point near the power tx, but NOT on a mounting bolt of that transformer!!! The grounds for the PI, powertubes, power stage filter E-caps, OT, filaments, power TX

3. Preamp stage ground terminal. On the furthest other side of the chassis,  next to the instrument inputs. I daisy chain the pot lugs that go to ground and ground them with 1 wire to that terminal. On the board I make a ground buss where all preamp grounds are connected to, including the E-caps, filtering the preamp B+ nodes. That ground buss is connected to the input grounding terminal  with a single wire. That's it..

[Mike_J]

My grounding is simple and effective:

1. grounding point dedicated to ONLY the power cord from the wall. As near as possible to the entrance point of the wall power into the chassis. The connecting wire should be longer than the L an N wires in case you rip that wire out of the chassis so ground wire is the last one to tear loose.

2. Power stage ground terminal. A star ground point near the power tx, but NOT on a mounting bolt of that transformer!!! The grounds for the PI, powertubes, power stage filter E-caps, OT, filaments, power TX

3. Preamp stage ground terminal. On the furthest other side of the chassis,  next to the instrument inputs. I daisy chain the pot lugs that go to ground and ground them with 1 wire to that terminal. On the board I make a ground buss where all preamp grounds are connected to, including the E-caps, filtering the preamp B+ nodes. That ground buss is connected to the input grounding terminal  with a single wire. That's it..

Like your system. Do you isolate your input jacks? Do you ground the presence pot with the pi on the power side or with the rest of the preamp grounds?

[thetragichero]

not to speak for someone else, but i'd bet the jacks are isolated. presence pot is basically the pi tail resistor so would be grounded with the rest of that section

[Mike_J]

I just choose a convenient point close to the PT and install a terminal strip.

I was just looking at http://sluckeyamps.com/misc/AB_Lite_II.pdf and you can see the grounding scheme as plain as day.

Very simple, and right to the point.

As I see it the 47 ohm resistor takes the place of the presence pot in the 5f6a. His initial pi tail resistor is much larger than the 10K in the 5f6a. He runs the ground wire from his version of a presence pot to the ground points for the bias section then to his main power ground. This treatment is logical to me. All is related to the pi as I interpret it.

[Mike_J]

not to speak for someone else, but i'd bet the jacks are isolated. presence pot is basically the pi tail resistor so would be grounded with the rest of that section

By grounded with the rest of that section are you saying the presence pot ground should be run to the bias grounds and on to the power side main ground?

[Mike_J]

I just choose a convenient point close to the PT and install a terminal strip.

I was just looking at http://sluckeyamps.com/misc/AB_Lite_II.pdf and you can see the grounding scheme as plain as day.

Very simple, and right to the point.

Sluckey also grounds his power tube cathodes on the same wire. This treatment is consistent with the Hoffman layout. The exception being the presence pot grounding to the preamp in the Hoffman layout.

[Mike_J]

not to speak for someone else, but i'd bet the jacks are isolated. presence pot is basically the pi tail resistor so would be grounded with the rest of that section

A significant factor at least from my point of view is the 5f6a has four input jacks. If they are grounded and I am looking for a place to go to ground I might get a little confused. Lift all jacks and run their grounds to one point on I guess the input side and I as an electron have no decision to make. Only one way to go.

[Mike_J]

The principle of the star point is that all 0V returns in that system connect back only to there. The only exception being with 2phase rectification; there the winding’s CT should connect directly to the reservoir cap terminal.

You have given me instruction on how to paint like Van Gogh. I am at paint by numbers.


I spent some time reviewing Aiken’s writing on ground points where I gleaned the following:
 
If you were to use three-star ground points the first would be the power supply and output stage ground and it would be located nearest to one edge of the chassis at the first power supply capacitor ground point.  The power grounds are the PT CT and the first filter cap ground. The output stages grounds include the output tube cathodes and OT secondary ground.
 
The second ground point might be used for the phase inverter circuitry and should be located at the ground point of the second filter capacitor (or the one that feeds the phase inverter), and this capacitor ground connection should be physically located upstream from the first star ground point. The third ground point would then be used for the remaining preamp circuitry and should be physically located upstream from the second star ground point.
 
Note that there still must be a ground return path to the rest of the circuit if the amp uses global negative feedback.  This should be in the form of a wire from the sleeve (or ground) connection of the output jack to the preamp ground point where the phase inverter common connections are grounded, or to the ground of the portion of the circuit where the global feedback from the tip connection of the output jack is connected. Merlin wrote, "If global feedback is used then the speaker ground should be returned to the local star of whichever stage the feedback happens to be applied to, which is usually the phase inverter

My first question is when Aiken says the phase inverter circuitry should be located at the ground point of the second filter capacitor (or the one that feeds the phase inverter) does he mean the ground point for the second stage cap that feeds the screen supply? Believe this was sufficiently answered by Merlin.


Second question has to do with the OT secondary ground. This is a 5f6a rebuild so it has negative feedback. Based on my reading as to what global negative feedback is I am pretty sure the 5f6a meets the criteria. Still confused as to whether the OT secondary ground should be attached to the first stage ecap ground point or the ground tab of the presence pot? Again answered by Merlin. Both the OT ground and the negative ecap ground should be connected to the ccw tab of the presence pot. The presence pot should not be grounded to the buss bar connecting the preamp pot grounds. The preamp should be grounded near the farthest input jack.


Thanks
Mike



Thanks
Mike

[pdf64]

From what I can make out, yes to both.
Consider that the OT secondary connection to 0V is purely as a reference, no current flows, so it’s not an 0V return per se.
You may find Merlin’s document on the topic helpful
http://www.valvewizard.co.uk/Grounding.pdf
Apologies I should have made you aware of it previously.

[Mike_J]

From what I can make out, yes to both.
Consider that the OT secondary connection to 0V is purely as a reference, no current flows, so it’s not an 0V return per se.
You may find Merlin’s document on the topic helpful
http://www.valvewizard.co.uk/Grounding.pdf
Apologies I should have made you aware of it previously.

I read Merlin's grounding document and ironically the part of it that most informed was his comment concerning connection to the main supply ground bolt. I believe his comment was  The safety-earth bond should be made to a dedicated screw close to the mains inlet. To be safe it must be completely impossible for the metal chassis (and anything else the user might touch) to become live. Once the chassis is earthed it will be at the same potential as the person using it, and if any live wire were to touch the chassis it would immediately be shorted to earth and cannot shock the user. Where the mains cable enters the chassis a heavy-gauge wire should be soldered to the earth tab and then connected to the chassis with a solder tag. The chassis area should be cleaned with emery paper. A nyloc nut should be used, or else a shake-proof or star washer should be used, with two ordinary nuts, well tightened.


Would you find the cleaned with an emery paper, bolt attached with a nyloc nut, or maybe a keps nut with an additional nut, well tightened attached to a solder tag an appropriate remedy for your to aversion to using the PT ground bolt?


So I am going to attach the OT ground to the presence pot ground tab and then send it to the second stage cap on the negative side. The grounding issue has been solved. The LTP ecap has nothing grounded to it now but if we weren't going to ground to the second stage then it wouldn't have anything grounded to it. Have to pick one or the other.


Thanks
Mike

[Mike_J]

Next question is an Aiken one.  He says "When using isolated jacks, be sure to connect a 0.01uF capacitor from the ground lug of the jack to the chassis with as short a lead as possible, to prevent radio-frequency interference (RFI)."

Now I thought I took care of this with the 68K grid stopper resistors. What is the deal here? If I am forced to do what he says, I have four jacks with their grounds all connected together. Can I at least just use one cap from the closest jack to the ground tab since they are all connected together? Another question, does this cap replace the wire I was intending to run to the tab anyway?

Thanks
Mike

[pdf64]

...
Would you find the cleaned with an emery paper, bolt attached with a nyloc nut, or maybe a keps nut with an additional nut, well tightened attached to a solder tag an appropriate remedy for your to aversion to using the PT ground bolt?

So I am going to attach the OT ground to the presence pot ground tab and then send it to the second stage cap on the negative side. The grounding issue has been solved. The LTP ecap has nothing grounded to it now but if we weren't going to ground to the second stage then it wouldn't have anything grounded to it. Have to pick one or the other.
...

As you may recall from your Aiken credibility thread, a PT fastener will tend to become loose over time, and additionally will be mechanically stressed, so isn't suitable for safety grounding the chassis.
No other chassis connection lugs are safety critical, so however unsuitable the PT fasteners might be double purposing, the likely worst case failure mode should only ever result in a reliability issue.
But aren't there any other fasteners you could use?

Fig 15.14 of Merlin's document shows the OT secondary referenced to the LTP 0V point.

[Mike_J]

...
Would you find the cleaned with an emery paper, bolt attached with a nyloc nut, or maybe a keps nut with an additional nut, well tightened attached to a solder tag an appropriate remedy for your to aversion to using the PT ground bolt?

So I am going to attach the OT ground to the presence pot ground tab and then send it to the second stage cap on the negative side. The grounding issue has been solved. The LTP ecap has nothing grounded to it now but if we weren't going to ground to the second stage then it wouldn't have anything grounded to it. Have to pick one or the other.
...

As you may recall from your Aiken credibility thread, a PT fastener will tend to become loose over time, and additionally will be mechanically stressed, so isn't suitable for safety grounding the chassis.
No other chassis connection lugs are safety critical, so however unsuitable the PT fasteners might be double purposing, the likely worst case failure mode should only ever result in a reliability issue.
But aren't there any other fasteners you could use?

Fig 15.14 of Merlin's document shows the OT secondary referenced to the LTP 0V point.

Yes, I agree neither Aiken nor Merlin believe the PT bolt is a suitable place for a ground point for the reasons you have mentioned. A better fastener than a keps nut supported by an additional nut. What could be better. Would probably use an 8-32. If anyone knows anything better please advise because I don't and I like fasteners. Main difference between ape and man in my opinion.


To get a little serious. Sorry I missed Merlin's Fig. Copied both Aiken and Merlin's ground documents onto Word so I could blow them up to a size that was easier for me to read. Took the Figs out when I did that. Will review Merlin's ground site you referenced again immediately.


Thanks
Mike

[Mike_J]

No excuses, it was in the text as well. Merlin wrote, "If global feedback is used then the speaker ground should be returned to the local star of whichever stage the feedback happens to be applied to, which is usually the phase inverter (e.g., fig. 15.14)." So it looks like the screen supply ground will be all by its lonesome unless I apply DC to the heater wires. One of them had to be. Will adjust the early reply to reflect this.

Thanks
Mike

[PharmRock]

Would probably use an 8-32. If anyone knows anything better please advise because I don't and I like fasteners.

In my current build (50W Marshall clone) I am using a dedicated #8 bolt with a keps nut.  I am using Hoffman's approach (listed in his library of info section) of crimping and then soldering a ring terminal to each ground wire and then attaching the ring terminals to the bolt.  My bolt is in the neighborhood of my PT and filter cap cans. 

A dedicated bolt is used for the mains ground, close to the entry point for AC wall power.  Similar approach...crimp the solder a ring terminal. 

Next question is an Aiken one.  He says "When using isolated jacks, be sure to connect a 0.01uF capacitor from the ground lug of the jack to the chassis with as short a lead as possible, to prevent radio-frequency interference (RFI)."

Now I thought I took care of this with the 68K grid stopper resistors. What is the deal here? If I am forced to do what he says, I have four jacks with their grounds all connected together. Can I at least just use one cap from the closest jack to the ground tab since they are all connected together? Another question, does this cap replace the wire I was intending to run to the tab anyway?

Thanks
Mike

Further in Aiken's tech info on this section he states the following:
"Important note: While the star ground is excellent for eliminating ground loop hum, it is not always the best scheme for preventing radio-frequency interference (RFI).  Fortunately, there is a simple addition to the star ground scheme that will make for a very quiet amplifier with no RFI.  Simply add a 0.01uF capacitor from the chassis to the ground lug of the isolated input jack using very short leads.  This will shunt all of the RF "riding" on the shield of the cord straight to the chassis ground before it can get into the amplifier and cause problems.   Any kind of ground lug can be used for the chassis connection of the capacitor.  You may be able to find a solder lug that slips over the shaft of the isolated input jack for a convenient ground lug.  Use of an internal-toothed lockwasher is recommended for these types of connections to insure a good "bite" into the chassis for a good ground."

If using Cliff-style jacks, TubeDepot sells a handy grounding washer for this at $0.25 each: https://www.tubedepot.com/products/grounding-washer

[Mike_J]

Would probably use an 8-32. If anyone knows anything better please advise because I don't and I like fasteners.

In my current build (50W Marshall clone) I am using a dedicated #8 bolt with a keps nut.  I am using Hoffman's approach (listed in his library of info section) of crimping and then soldering a ring terminal to each ground wire and then attaching the ring terminals to the bolt.  My bolt is in the neighborhood of my PT and filter cap cans. 

A dedicated bolt is used for the mains ground, close to the entry point for AC wall power.  Similar approach...crimp the solder a ring terminal. 

Next question is an Aiken one.  He says "When using isolated jacks, be sure to connect a 0.01uF capacitor from the ground lug of the jack to the chassis with as short a lead as possible, to prevent radio-frequency interference (RFI)."

Now I thought I took care of this with the 68K grid stopper resistors. What is the deal here? If I am forced to do what he says, I have four jacks with their grounds all connected together. Can I at least just use one cap from the closest jack to the ground tab since they are all connected together? Another question, does this cap replace the wire I was intending to run to the tab anyway?

Thanks
Mike

Further in Aiken's tech info on this section he states the following:
"Important note: While the star ground is excellent for eliminating ground loop hum, it is not always the best scheme for preventing radio-frequency interference (RFI).  Fortunately, there is a simple addition to the star ground scheme that will make for a very quiet amplifier with no RFI.  Simply add a 0.01uF capacitor from the chassis to the ground lug of the isolated input jack using very short leads.  This will shunt all of the RF "riding" on the shield of the cord straight to the chassis ground before it can get into the amplifier and cause problems.   Any kind of ground lug can be used for the chassis connection of the capacitor.  You may be able to find a solder lug that slips over the shaft of the isolated input jack for a convenient ground lug.  Use of an internal-toothed lockwasher is recommended for these types of connections to insure a good "bite" into the chassis for a good ground."

If using Cliff-style jacks, TubeDepot sells a handy grounding washer for this at $0.25 each: https://www.tubedepot.com/products/grounding-washer

I see nothing wrong at all with Hoffman's method of crimping and soldering a ring terminal as sound practice #8 bolt and a keps nut. Merlin suggests using an additional nut tightened down firmly and I don't see how that can hurt.


As far as using the .01 uF cap from jack ground lug to chassis still not sure if that is supposed to replace the ground wire which would normally go from the isolated jack to the chassis ground.

[pdf64]

...As far as using the .01 uF cap from jack ground lug to chassis still not sure if that is supposed to replace the ground wire which would normally go from the isolated jack to the chassis ground.

The cap is additional to the actual 0V common connection. Good practice if its / their chassis 0V connection is not already close by the input jack/s.

[Mike_J]

...As far as using the .01 uF cap from jack ground lug to chassis still not sure if that is supposed to replace the ground wire which would normally go from the isolated jack to the chassis ground.

The cap is additional to the actual 0V common connection. Good practice if its / their chassis 0V connection is not already close by the input jack/s.

Thanks for the good information. Jack and ground point won't be farther than two inches away from each other. Will try without. Have plenty of .01uF caps if I start hearing songs that aren't coming from my guitar.

[Mike_J]

I just choose a convenient point close to the PT and install a terminal strip.

I was just looking at http://sluckeyamps.com/misc/AB_Lite_II.pdf and you can see the grounding scheme as plain as day.

Very simple, and right to the point.

Merlin has me going in another direction. Suggested what he calls the bus bar system which only grounds the board, pots, PT, OT and input jacks at one place and that is the input jack. In my case with isolated jacks it will be a tab bolted to the chassis near the input jacks.


The Hoffman layout system lends itself very well to this idea. While it isn't entirely a bus bar system it essentially performs the same function with a little modification. Merlin said it makes for a quiet amp but mentioned figuring out where to ground the heater supply may be a challenge. I decided just to lift it by connecting it to the heater DC supply. That way I don't have to figure out where to ground it because the DC supply will do it. May still be a problem but hopefully not.