Balanced Power Supply Filtering

Discussion in 'Shock Brother's DIY Amps' started by SerpentRuss, May 6, 2021.

  1. SerpentRuss

    SerpentRuss TDPRI Member Ad Free Member

    Age:
    58
    Posts:
    98
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    In Merlin Blencowes's preamp book, in the grounding chapter, he talks about using balanced RC filtering in the power supply to reduce noise. This would create a new ground reference after the power supply ground that would be elevated a few volts.

    As anyone tried this or does anyone know of any designs that incorporate this concept? It would require isolating pots from the chassis and if used more than once, at each dropping resistor, carefully grounding each stage of the amplifier to the proper location.

    I am throwing a basic schematic up here, as an illustration, the values are random, so don't pay any attention to them.

    Balanced PS Filter.JPG
     
  2. Tom Kamphuys

    Tom Kamphuys Tele-Holic

    Age:
    41
    Posts:
    753
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    I have not tried it. Couldn't get my head around it.

    You don't need to isolate your pots. My humdinger pot is elevated 60V or so. Housing is not connected to any of the lugs. I explicitly checked it.
    No need to carefully ground at each dropping resistor. Only ground at the input, just like Merlin says.
    I'm not sure if you can use the balanced part more than once; there is a situation in which you can use it, but I wasn't sure if that was the case in my amps, not even once.

    Safety ground is the most important ground. It is directly from the 'mains cable' to the chassis, so before the transformer and thus the rectifier, and is not part of the circuit
     
    JuneauMike, SerpentRuss and sds1 like this.
  3. JuneauMike

    JuneauMike Poster Extraordinaire

    Posts:
    6,977
    Joined:
    May 5, 2015
    Location:
    Alaska
    Subscribed

    @SerpentRuss, it would be cool to scale this down for 9 volts and breadboard it out. Do you have the stuff to do that? (Although, I'm not sure how you'd quantify noise in the signal chain, which is the whole point of this exercise).
     
  4. Lowerleftcoast

    Lowerleftcoast Friend of Leo's

    Posts:
    3,670
    Joined:
    Dec 30, 2019
    Location:
    california
    I don't have any recollection of any design that uses this but I haven't seen everything yet. I have considered trying this even though designs without additional resistance/impedance tend to be quiet.
    I am a student of this as well, but from my understanding, you have this wrong.
    Merlin Blencowe wrote: "In general we would try to minimise the ground impedance, but there is a useful exception to this rule." He also wrote:"If there is an additional stage of smoothing immediately after the reservoir capacitor then it is beneficial to split the dropping resistance into two parts to create a balanced filter..." I did not read a suggestion to replace the ground impedance at any other location to create balanced filters.

    Your diagram does not show the additional stage of smoothing immediately after the reservoir capacitor. It also shows more than one reference to ground. The only place Merlin suggests for the ground is near the input.

    Merlin also wrote: "One or both of the dropping resistors could be replaced by a smoothing choke, and there is the added advantage that a choke in the negative side of the circuit is subject to less voltage stress between the coil and frame than one in the ‘traditional’ side of the circuit."
    I find this interesting. Evidently this smoothing choke not does have to equal the choke on the B+ rail. Perhaps a simple air coil would be beneficial. I plan to install a simple air coil on my next SE build in this location.
     
  5. SerpentRuss

    SerpentRuss TDPRI Member Ad Free Member

    Age:
    58
    Posts:
    98
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    C2 in my diagram would be the "Additional stage of smoothing" and my Amplifier ground would not be at the same potential as earth/safety/PS ground. So even though I'm showing two grounds, the amplifier ground is not "earthed"
     
  6. SerpentRuss

    SerpentRuss TDPRI Member Ad Free Member

    Age:
    58
    Posts:
    98
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    I was thinking about trying this with a pipsqueak-sized amplifier, 6N2P Pre-amp, 6AK6 SE output. I just wanted to hear some opinions about it first.
     
    JuneauMike likes this.
  7. JuneauMike

    JuneauMike Poster Extraordinaire

    Posts:
    6,977
    Joined:
    May 5, 2015
    Location:
    Alaska
    My only concern, which comes mostly from ignorance, is that you would be using mains power instead of a reduced voltage. We lick 9v batteries to see if they are still good. So that just seems way more safe then potentially sending wall voltages through a capacitor without a blast shield in place. ;)
     
  8. SerpentRuss

    SerpentRuss TDPRI Member Ad Free Member

    Age:
    58
    Posts:
    98
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    You're right. I keep on thinking the post and housing are connected to one of the terminals, but they're not.
     
  9. SerpentRuss

    SerpentRuss TDPRI Member Ad Free Member

    Age:
    58
    Posts:
    98
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    Think of AC signal in my diagram as being the output of a fictional PT, not connected directly to the mains. Here is a trace trend from that circuit. The green trace is at C1, the blue trace is at C2 and the red trace is the amplifier group reference on the other side of the load resistor. You can see that since that ground is elevated and has the same ripple as the blue trace.

    Balanced Trend.JPG
     
    JuneauMike likes this.
  10. JuneauMike

    JuneauMike Poster Extraordinaire

    Posts:
    6,977
    Joined:
    May 5, 2015
    Location:
    Alaska
    Ah, that's why we often solder ground through the lug and onto the pot casing itself. Because otherwise it would be isolated and not grounded? Duh.
     
  11. Lowerleftcoast

    Lowerleftcoast Friend of Leo's

    Posts:
    3,670
    Joined:
    Dec 30, 2019
    Location:
    california
    For C2 to be the "Additional stage of smoothing" R3 would have to be added. The red circle would be the next node.

    InkedBalanced PS Filter_LI.jpg
     
    2L man likes this.
  12. SerpentRuss

    SerpentRuss TDPRI Member Ad Free Member

    Age:
    58
    Posts:
    98
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    I think you're taking my circuit a bit too literally.:) Why did you cross out the text for the ground references, this is the point of the question I'm asking? They are a different potential.
     
  13. Lowerleftcoast

    Lowerleftcoast Friend of Leo's

    Posts:
    3,670
    Joined:
    Dec 30, 2019
    Location:
    california
    ??? I have paid no attention to values of the cap/resistors. Mainly I pointed out that the illustration did not show the additional stage of smoothing.
    First off, the illustration has one too many ground reference points attached to this power filtering circuit. Quoting Merlin: "There must be no connection between transformer-rectifier-reservoir and chassis, of course, but that is always true of a proper ground scheme."

    Secondly, the input end of the amplifier is not shown on the illustration. To show a ground reference point on the power tube node might throw off some of the future readers of this thread. Ime ground schemes are difficult enough to understand. Quoting Merlin again: "One logical ground scheme is the bus* ground. This is involves routing a single, heavy-gauge wire –called the bus wire or bus bar– through the chassis. The path of the bus wire should follow the natural path of the amp circuit from the reservoir capacitor, to power amp (if present) to preamp, to input stage, and all ground connections are made progressively along it in the same natural order. The bus bar should be connected to the chassis at one point only, at the input end of the amplifier..."
     
  14. SerpentRuss

    SerpentRuss TDPRI Member Ad Free Member

    Age:
    58
    Posts:
    98
    Joined:
    Feb 17, 2021
    Location:
    North Carolina

    I truly think you're misinterpreting or overthinking this. The text indicates the amplifier ground is not earthed, it is floating, not connected to the chassis. I should have used an arrow and not an electrical connection (dot). Your drawn-in R3 would be rolled into the rest of the circuit and can be represented by RLoad. C2 is without a doubt functioning as a filter capacitor. As for a ground bus, that would theoretically be the trace from the base of C2 to the load resistor. As for the ground at C1, I'm not showing how it's wired, just the fact that it's earthed, as it is in any safely built amplifier.
     

    Attached Files:

    Last edited: May 6, 2021
  15. Lowerleftcoast

    Lowerleftcoast Friend of Leo's

    Posts:
    3,670
    Joined:
    Dec 30, 2019
    Location:
    california
    I have been known for that. lol. I will try to catch up as the thread progresses.
    R3 would not be represented by RLoad. Rload is the 6AK6 power tube. R3 is the dropping resistor between C2 and C3*. R3 also forms an RC filter with C3.

    * C3 is the capacitor drawn within the red circle shown in post#11.
     
  16. Ten Over

    Ten Over Tele-Holic

    Posts:
    800
    Joined:
    May 13, 2015
    Location:
    Central California
    Maybe a different picture will help. The balanced reservoir cap is only for power supplies with two stages of filtering for the plates.
    Power Supply Merlin Balanced Reservoir.png
     
    Tom Kamphuys likes this.
  17. SerpentRuss

    SerpentRuss TDPRI Member Ad Free Member

    Age:
    58
    Posts:
    98
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    I see! I saw this diagram in the book and it shows leads "from the rectifier." I assumed the chassis ground reference was there. My apologies for being so thick-skulled. Have you tried this?
     
    Lowerleftcoast likes this.
  18. Tom Kamphuys

    Tom Kamphuys Tele-Holic

    Age:
    41
    Posts:
    753
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    Very clear post.

    Are there any amps with two stages? The ones I've checked don't.
    I guess it will influence sag and recovery from it, right?
     
  19. 2L man

    2L man Tele-Meister

    Age:
    61
    Posts:
    455
    Joined:
    Nov 23, 2020
    Location:
    Finland
    I built that kind of PS without Choke to my latest SE which is Princeton with MV and FB pots and uses 5881 and it is my quietest SE but it is low gain amp. But it quieter than previous Princeton 6V6 which has a Choke.

    There are five RCR stages before OPT output. First four 16uF electrolyts and fifth is 4,7uF poly so eight resistors between capacitors. Cost still less than a choke and did fit to a turrett board :)

    And like all my builds I don't use chassis as active amplifier circuit and all return currents flow to a single ground bus.
     
  20. 2L man

    2L man Tele-Meister

    Age:
    61
    Posts:
    455
    Joined:
    Nov 23, 2020
    Location:
    Finland
    Ground bus or any components should not be soldered to potentiometer cases when they are mounted to metal chassis because there comes a possibility to form ground loops!!!

    Especially when chassis is used to flow return current.

    Now when no current flows thru pot case the pot screw and star washer makes good enough electrical connection and pot case as a part of chassis forms a shield over rather high impedance tube amp component where on pre amp volume signal is still lowish and noise there comes will be amplified a lot on following stages.
     
    Last edited: May 7, 2021
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.