Need help on 5E2 filament center tap and what to ground

KeithAmpsIt

TDPRI Member
Joined
Aug 15, 2022
Posts
21
Age
60
Location
Olathe, KS
I am starting to build a 5E3 amp, my first build. I am not sure how to wire up the green/yellow center tap from the PT for the 6.3V filaments. The Fender schematic doesn't help much. I saw one layout that used two 100 ohm resisitors across the lamp then to ground. I have made my own schematic layout (see attached) so I can understand how things actually wire up. When I did that I see a problem that the green/yellow center tap wire is shown grounded and also from the two resistors. I'm thinking that maybe I don't need those two resistors? Am I showing the correct location for the green/yellow center tap?
Any help is appreciated!
 

Attachments

  • Fender 5E2 schematic.pdf
    1.5 MB · Views: 26
  • 5E2 Schematic & Layout.pdf
    1.8 MB · Views: 30

chas.wahl

Tele-Holic
Joined
Dec 8, 2012
Posts
595
Location
NYC
If you have a green-yellow center-tap wire, then you can ground that. The two 100-ohm or thereabouts resistors are for making a "virtual" center tap to establish ground reference if your PT doesn't provide center tap.

Merlin Blencowe (aka Valve Wizard) has observed that a lot of times the center tap is not really balanced, or even if it is, something about the heater system makes it unbalanced enough that it causes hum anyway, even with a center tap. His recommendation, to "dial out" hum to the greatest extent, is to use a 250-500 ohm pot to replace the center tap, or the pair of resistors. The two legs of the heater circuit are connected to the outer lugs of the pot, and the wire to ground is connected to the middle lug (the wiper). You can read about this in his Heater Supplies article: https://www.valvewizard.co.uk/heater.html (where he also explains about the virtual center tap).

The other thing to consider is where to connect that ground reference; in the same article Blencowe suggests that elevating the heaters on a mild DC voltage (such as the cathode of one of the power tubes) will also help reduce hum.

If you really want to take a deeper dive, you might also check out his Grounding article.
 

Lowerleftcoast

Poster Extraordinaire
Joined
Dec 30, 2019
Posts
6,023
Location
california
Welcome to TDPRI.

This thread may be of interest. It is for a 5F2A but I am sure you can glean quite a bit for your build.
 

KeithAmpsIt

TDPRI Member
Joined
Aug 15, 2022
Posts
21
Age
60
Location
Olathe, KS
Thanks for the help! Ive updated my schematic to correct the filament ground issue and will get it grounded over to the Pin 8 of the 6V6GT. See what you think. I will look more closely at the thread Lowerleftcoast just posted. One thing I've seen already is I may want to put in a bleeder resistor across my first filter cap. Is this resistor mainly for a safer environment that allows the caps to discharge more readily once the power is turned off?
 

Attachments

  • 5E2 Schematic & Layout rev1.pdf
    781.5 KB · Views: 18

chas.wahl

Tele-Holic
Joined
Dec 8, 2012
Posts
595
Location
NYC
The bleeder cap "automatically" dissipates the charge in the first and other connected filter caps (those other than the one bridged via intervening dropping resistors). It works continuously, so there is some power loss through it while the amp is in operation -- the amount dependent on the value of the bleed resistor and the voltage feeding it.

Unless you are willing to install such resistors on all the amps you own, one consideration is whether you'll remember which amps have this feature, and which don't. In some respects, Always Remembering to use a separate, dedicated bleed fixture with well-insulated clips might be a better idea. Pros and cons to every decision . . .

You've obviously drawn your schematic with the "split circuit ground" scheme that most builders seem to use. I will only note that both Merlin Blencowe, and more lately even Rob Robinette, have proposed that the circuit ground reference be provided at only one point, at or near the input jack. See Blencowe's Grounding chapter and Robinette's "Optimized 5E3 Layout" -- while the 5E3 is a bit more complicated, the basic idea is shown clearly there. That said, a lot of people say that the split ground can be quiet too. I'm not an electron or a guitar signal, so I don't have a vote.
 
Last edited:

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
8,755
Location
Salt Lake City
Welcome. A 5E3 is a great build, and you're smart to do a lot of schematic thinking and layout drawing or doodling.

I'll echo the endorsement by @chas.wahl for Merlin's heater chapter. Short clear, relevant. (His grounds chapter is good too, but long, technical, and sometimes less clear for non-wizards. It also mixes important ideas with more theoretical ones. I re-read it twice a year and now claim to understand half of it. :))

By all means keep drawing and marking up diagrams. The Fender version is a great starting point. For the 5E3, Rob's standard layout is just about perfect, and has been built hundreds of times with great results. He includes a lot of important ideas.

5e3_DIYLC.png


For those who insist on a single circuit ground (the safety ground must always be separate), Rob's optimized layout is great but requires a custom board. FWIW, I drew a single-ground on a stock board a while back. *But* I don't push folks away from Rob's split ground.

Rob 5E3 1 ground 0 standby.png
 

chas.wahl

Tele-Holic
Joined
Dec 8, 2012
Posts
595
Location
NYC
Erm: He's building a 5E2. I only suggested the 5E3 layout because it shows the single ground reference graphically.
 

KeithAmpsIt

TDPRI Member
Joined
Aug 15, 2022
Posts
21
Age
60
Location
Olathe, KS
Erm: He's building a 5E2. I only suggested the 5E3 layout because it shows the single ground reference graphically.
I misspoke earlier. As Chas.wall says I am actually building a 5E2. Sorry for any confusion. I am planning on grounding the signal side at the input jacks.
 

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
8,755
Location
Salt Lake City
Aha, I trusted your first sentence, shoulda trusted your title and all your nice diagrams, which also show you know your way around quite well. Sorry.

I hope some of the 5E3 material applies. Just in case it might help, here's a split-bus 5F2a I drew a while back -- closer, if no cigar? You may have it in mind, but note the HT CT to the first filter negative cap.

5F2a modern simple.png
 
Last edited:

KeithAmpsIt

TDPRI Member
Joined
Aug 15, 2022
Posts
21
Age
60
Location
Olathe, KS
No problem. Ive been looking at a lot of the similar amp schematics and layouts. I started out to build a 5F2A until I changed to using a choke for smoothing the DC supply. I'm seeing that every little change gets it own name! On your layout KingFan, excuse my ignorance but how is that different than the HT CT going straight to the power amp ground lug if the black jumper stays like you have shown?
 

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
8,755
Location
Salt Lake City
how is that different than the HT CT going straight to the power amp ground lug if the black jumper stays like you have shown?

Yeah, I didn't understand it at first either. For theory, read Merlin on Ground, very specifically section 15.3. That's not 100% intuitive, either; let me try this. We want to provide as straight a shot as possible back to the PT for that big noisy rectifier ripple return current coming out of the reservoir cap. We don't want to put it on the ground bus. The old way, the ripple could get back two ways, left or right, on a split bus (and *only* via the bus on a single bus)

1661131258354.png


The new way creates a tighter, more direct loop from rectifier through cap back to PT.

1661131283715.png


Final note: When I did this on my 5F2a, the reduction in hum was easily audible. And it couldn't be simpler.

EDIT: To be fair, that 5F2a has a single bus. I hadn't thought of it before, but is this HT CT trick even more important when there's not the power amp ground nearby? I'm not clever enough to say. Either way, the HT CT trick was also quieter on my split-bus 5E3.
 
Last edited:

chas.wahl

Tele-Holic
Joined
Dec 8, 2012
Posts
595
Location
NYC
You're right about Merlin Blencowe not stating making the above conclusion very obvious. It doesn't help that he sometimes orients his diagrams with power section on the left (like in a Fender layout, but upside down) and sometimes with power section on the right (as in a Fender schematic) -- and the diagrams are sometimes drawn like layout, and sometimes like schematics, or a mix of both.

Does putting a resistor in the ground bus between the filter cap and the next one help to confine the ripple to the center tap (MB's "balanced filter" shown in Fig 15.7) -- or is that indeed the whole point of it? I never really understood this part of this chapter, but maybe I'm getting closer.
1661193650546.png
 
Last edited:

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
8,755
Location
Salt Lake City
Aha, good one. I never put 2 and 2 together there -- as you say, that resistor is in the ground chain, and he says "will retard ripple current ... entering the audio ground." I think I've seen @Lowerleftcoast discuss this approach; I'll bet he knows.

Merlin, of course, is basically always talking about a single audio ground. Smarter folks than I will know if keeping ripple off a single ground is even more important than keeping it off a split 'power amp' ground. That was yesterday's 'aha' moment -- but I'm still just speculating.
 
Last edited:

KeithAmpsIt

TDPRI Member
Joined
Aug 15, 2022
Posts
21
Age
60
Location
Olathe, KS
I've been reivewing all the things you guys have said. Lots of things to digest. Going back to King Fan showing the 6.3V CT (green-yellow) connecting to the cathode pin 8 to elevate the ground for the filaments.

Something is puzzling me about this. You want to connect to a point of constant voltage, right? I looks like to me that connecting the CT there would not be a point of constant voltage. I think its only constant there during idle. During operation, the cathode voltage could vary from 0V to ~35V (saturation to cutoff). Isn't the whole point of an elevated center tap for the filaments to have a reliable constant reference voltage? Let me know if I'm wrong about this.
 

chas.wahl

Tele-Holic
Joined
Dec 8, 2012
Posts
595
Location
NYC
The point of this is to elevate the AC heater on a DC base, the cathode voltage. It apparently doesn't matter, within reason, whether that DC voltage fluctuates some. I suggest that if you haven't already, you read Merlin Blencowe's chapter on Heaters.
 

KeithAmpsIt

TDPRI Member
Joined
Aug 15, 2022
Posts
21
Age
60
Location
Olathe, KS
I did read it to the best of my understanding. :) That voltage fluctuation would alter the 6.3 v ac quit a bit I’d think. Granted I don’t have experience having actually done this like you have.
 

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
8,755
Location
Salt Lake City
Floating an AC current on a DC voltage doesn't alter the size of the AC signal at all. You're waving your arm up and down while standing on a boat in the harbor. The tide goes up and down under the boat. The arm-waving amplitude doesn't change.
 

2L man

Tele-Afflicted
Joined
Nov 23, 2020
Posts
1,785
Age
62
Location
Finland
I've been reivewing all the things you guys have said. Lots of things to digest. Going back to King Fan showing the 6.3V CT (green-yellow) connecting to the cathode pin 8 to elevate the ground for the filaments.


Something is puzzling me about this. You want to connect to a point of constant voltage, right? I looks like to me that connecting the CT there would not be a point of constant voltage. I think its only constant there during idle. During operation, the cathode voltage could vary from 0V to ~35V (saturation to cutoff). Isn't the whole point of an elevated center tap for the filaments to have a reliable constant reference voltage? Let me know if I'm wrong about this.
Yes you are right! Max signal amplitude on cathode biased power tube cathode depends of bypass capacitor value and typical 10uF I recall comes about 5Vpp at 80Hz. After I saw that on oscilloscope I quit using it as filament elevate although I did not do any measuring whether it have any adverse effect.

I usually install a bleed circuit which first has a 220k resistor and series to Common with it parallel 56k resistor and 10uF 100V electrolyt where I connect filament CT. Then depending which B+ the bleed is connected and its voltage the filaments elevate to 20% which usually comes somewhere 50V to 80V. When amp does not have cathode follower(s) bottom resistor can be 47k.
 

chas.wahl

Tele-Holic
Joined
Dec 8, 2012
Posts
595
Location
NYC
Maybe think of it this way: the whole heater system is operating at an elevated voltage level, meaning the "ground potential" is higher than everywhere else on earth (zero volts or thereabouts). The power for the heaters, then, is developed by the difference in potential between the legs of the heaters and that ground reference.
 




Top