beta22 Build

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Because I still have more fingers on my hands than I have projects on-the-go, I thought I'd start another, and overwhelm myself! :banana:

So, for a looong while, my favourite downtown surplus shop has had this beautiful blue Leitch rack-mount gear kicking around; I thought this one (an FR-664 Distribution Amplifier) had the looks to serve as a home for my headphone amp build. The corners (small to large radius corners) and grill were what caught my eye. The face is cast and slides out to open up--I'm using this feature to my advantage, hiding the plugs and switches behind the grill to maintain the cool looks when it's not in operation, then just flip the front down to turn on and turn up!

IMGP0376_sm.jpg




IMGP0365_sm.jpg






And here's a proposed layout that I've got in my mind. I plan to have a barrier between the sigma22 + transformer board and the epsilon22 backplane, to separate the power and audio sections. (possibly one piece of steel, or two piece with an air gap between them for isolation)


IMGP0411_sm.jpg


IMGP0398_sm.jpg
 
sorry to see you've bought the "3-channel" hype

in that space dual mono, dual polarity supplies should easily fit

with split bobbin EI xfmrs you may well want some iron between the ps and amp boards but the line rejection, channel separation and "gnd contamination" performance would be much better than "3-channel" with a single toroidial xfmr supply

could be cheaper than stuffing a 3rd amp board too

you can search for my (jcx) posts on head-fi with "3-channel" and "active gnd" for an engineering based critique of "3-channel"'s failings
 
Last edited:
I didn't buy into any hype. I bought the b22 board, e22 backplane, plus all matched transistors, sinks, CCSs, etc. including parts for the s22 board (had to buy the board separately) for $cheap, shipped. Thanks for the suggestions, but I'm doing what I can within my means :)

All I've left to buy are resistors and caps, and I'm buying them in five-and-ten quantities anyways, so the extra cost for the third channel is virtually non-existent. When everything's done, I can put it in or take it out of circuit and see for myself which sounds better to me--thanks for the suggestion, you've got me thinking now.

I am using a Plitron dual secondary 30V 300VA transformer for the supply and I plan to condition the incoming 120V.

I hope to include binding posts for speakers and a I've plans for a switch circuit to flip-flop between speakers and headcans :)

I've read that the active ground can help with the reverse currents induced by speakers' spiders running the coil back thru the magnet. jcx, have you much to say on that subject?
 
Last edited:
I've read that the active ground can help with the reverse currents induced by speakers' spiders running the coil back thru the magnet. jcx, have you much to say on that subject?


Where'd you read that at?

se

That's what I gathered from AMB's site, under the b22 'Technical Highlights' section...

www.amb.org said:
3-channel "active ground" amplifier (3 β22 boards required)

This is the recommended configuration for standard 3-wire headphones, and offers improved performance by having an active ground channel amplifier for the headphone's shared "ground return" wire. The ground channel amplifier sources or sinks the return current from the transducers, which would otherwise have been dumped into signal ground or power supply ground. This shifts responsibility for the high current reactive load of the headphones from signal ground to the tightly regulated power supply rails, thus removing the primary source of signal ground contamination. The headphone transducer "sees" symmetrical output buffers with equal impedance and transfer characteristics on both sides, rather than an amplifier on one side and a capacitor bank of the power supply ground on the other. This results in lower output impedance, greater linearity and reduced stereo crosstalk.

The 3-channel configuration could also be used for driving speakers (the ground channel output should be connected to both the left and right speakers' negative terminals). The ground channel will bear the return current of both stereo channels and must therefore have larger heatsinking. A total of three σ22 PSUs is recommended for this configuration (one per β22 board).


Have I misinterpreted the purpose of the third channel from the description provided?

I'm trying to learn as I go, here... :)
 
Last edited:
Hi, I don't think you have misinterpreted, but I do think that you have 'bought some marketing'. Anyway that amp is well regarded so you are certainly going in the right general direction!

So, ground... most people here tend to regard the best ground as being ground. If you separate the left and right shared ground into two separate grounds, and keep them separate all the way back to the power supply 'star' point, then you will find really good performance from just two channels.

I am sure many headphone fans like to call it 'balanced', but is not; it is just normal separation like speakers. (You certainly do not need a bridged amp, which again many mistakenly call 'balanced').

You will need a 4-way socket, such as a 4 pole XLR (...or possibly 2 x 3 pole XLRs, but they are a bit big).

One layer of steel should work as a shield... but remember to ground it well.

Finally... invest in a good quality volume control with decent left / right matching.

Your case looks classically cool, so you have all the ingredients of a super little project... so the best of luck!

: )
 
Have I misinterpreted the purpose of the third channel from the description provided?

There was some misinterpretation on your part, but more important is the misrepresentation of that description on AMB's part.

It perpetuates the myth that an active ground causes return currents to bypass signal and power supply ground (which was started by a user named Morsel who got the whole "active ground" thing started).

The return currents flow through the signal and power supply grounds with an "active ground" just as they do with a conventional ground, only they take a more circuitous route getting there.

The "active ground" myth has people believing that the return currents either dribble out onto the floor or evaporate into the ether. They don't. An an "active ground" offers no particular benefit over a well-executed conventional ground.

I'm trying to learn as I go, here... :)

Hey, as long as you're trying to learn, you're in good shape! :)

se
 
Thanks Jen-B very much for your comments; that sounds like an interesting design, however I'd imagine a recable would be in order for my headphones to make use of such a layout.

Thank you for the explanation, Steve Eddy. I understand that the current has to flow back thru the circuit, but doesn't the active ground do anything in terms of bolstering and stabilizing the ground reference?

I think I will try to build the amp with the three channels, just to see what everyone else is talking about (I already have the expensive bits for it) and if I don't like it I'll just take it out, I don't think that should cause me any problems.

In any event, if I were to build this with XLR connector (/connectors, be it one 4-pin, or two 3-pin) plus a normal 1/4" jack, as Jen-B has outlined, would it be possible to implement a switch which would joint or separate the L and R channel ground just before they hit the jack? In essence, a switch that would require toggling to change between the XLR jack(s) and the 1/4" plug, or would I run into ground loop issues? (I don't think I would, but I figure I'd ask)

Diggin' this dialog we've got going on :)
 
Last edited:
Thank you for the explanation, Steve Eddy. I understand that the current has to flow back thru the circuit, but doesn't the active ground do anything in terms of bolstering and stabilizing the ground reference?

Nope.

The ground reference is ultimately nothing more than a singular point. A node in the circuit. There's nothing to "stabilize" or "bolster." And whether "active" or passive, all the return currents flow through that node.

I think I will try to build the amp with the three channels, just to see what everyone else is talking about (I already have the expensive bits for it) and if I don't like it I'll just take it out, I don't think that should cause me any problems.

Couldn't hurt.

In any event, if I were to build this with XLR connector (/connectors, be it one 4-pin, or two 3-pin) plus a normal 1/4" jack, as Jen-B has outlined, would it be possible to implement a switch which would joint or separate the L and R channel ground just before they hit the jack? In essence, a switch that would require toggling to change between the XLR jack(s) and the 1/4" plug, or would I run into ground loop issues? (I don't think I would, but I figure I'd ask)

No need for that. Since they all must come together at some point, you can simply run separate wires to that point, i.e. two wires from the XLRs and one wire from the TRS.

Since the B22 boards only have one pad for the output, you're not going to be able to tie them all to that point unless you're using some rather small gauge wire. So you might want to use a tag strip and bring everything together on one of the tags.

tagstrip2.jpg


Just make sure you don't use the tag that's tied to the chassis.

Diggin' this dialog we've got going on :)

Great! That's what forums like this are all about! Or SHOULD be about anyway. :D

se
 
In any event, if I were to build this with XLR connector (/connectors, be it one 4-pin, or two 3-pin) plus a normal 1/4" jack, as Jen-B has outlined, would it be possible to implement a switch which would joint or separate the L and R channel ground just before they hit the jack? In essence, a switch that would require toggling to change between the XLR jack(s) and the 1/4" plug, or would I run into ground loop issues? (I don't think I would, but I figure I'd ask)

No need for that. Since they all must come together at some point, you can simply run separate wires to that point, i.e. two wires from the XLRs and one wire from the TRS.

Since the B22 boards only have one pad for the output, you're not going to be able to tie them all to that point unless you're using some rather small gauge wire. So you might want to use a tag strip and bring everything together on one of the tags.

tagstrip2.jpg


Just make sure you don't use the tag that's tied to the chassis.

According to Jen-B's description of how the XLR output should be handled;

...
So, ground... most people here tend to regard the best ground as being ground. If you separate the left and right shared ground into two separate grounds, and keep them separate all the way back to the power supply 'star' point, then you will find really good performance from just two channels.
...

So for the TRS, I would run a wire from the star point back near the PSU for the ground pin, then use the output grounds on the b22 boards for the XLR connectors?

I'm familiar with tag boards--they're used a fair bit in tube amps. There's any number of ways to split the wiring--I could use screw down terminals and cram the wires in, or I could crimp two wires to a female pin in a plug if I solder on a header to the board, or just put a long pin at the output and wirewrap, etc... that decision can wait ;)



Alrighty, so a piece of 3/16" ~ 1/4" thick (chosen for ability to drill and tap the edge for mounting) steel plate should be fine for PSU hum isolation from the b22 boards if I put the whole shebang in one box, then notch an edge to let the power supply wires run through (rather than feeding thru a hole and crimping on the other side, for modularities' sake).

I gotta crack open the CAD suite and see what I can come up with...
 
According to Jen-B's description of how the XLR output should be handled;

That was assuming a two board build. Not a three board build.

So for the TRS, I would run a wire from the star point back near the PSU for the ground pin, then use the output grounds on the b22 boards for the XLR connectors?

Technically you could do that, but if you're going to start out with a three board build, why? Basically that would have your TRS outputs running like a two board build and the XLRs like a three board build.

If you're going to start with a three board build, why not run all outputs as a three board build?

I'm familiar with tag boards--they're used a fair bit in tube amps. There's any number of ways to split the wiring--I could use screw down terminals and cram the wires in, or I could crimp two wires to a female pin in a plug if I solder on a header to the board, or just put a long pin at the output and wirewrap, etc... that decision can wait ;)

Oh sure, there's more than one way to skin that cat.

Alrighty, so a piece of 3/16" ~ 1/4" thick (chosen for ability to drill and tap the edge for mounting) steel plate should be fine for PSU hum isolation from the b22 boards if I put the whole shebang in one box, then notch an edge to let the power supply wires run through (rather than feeding thru a hole and crimping on the other side, for modularities' sake).

Sounds like a plan.

I gotta crack open the CAD suite and see what I can come up with...

Good luck!

se
 
As SE noted, my comments were based on the notion that you might be swayed towards a two channel design. However if you are going three channel then stick with a single 3-pole TRS jack connector!

Hope it goes well.
J.

Thanks J!

I've got some 1/4" Neutrik TRS jacks that should do the job ;) I'll build in the ability for my to switch between two and three boards and we'll see which sound I like better before I button it up.

At the cottage all this week so taking a bit of a break for now!


:drink: :whip: :wrench: :wrench: :smash: :drink:





:nownow:
 
while I think a 3rd board for active gnd is wastefull my real argument is with "3-channel" specifically - using the active gnd "3rd channel" to "buffer" signal gnd so that it is in series with the signal/feedback gnd and the driver's (common)gnds - and outside of the feedback loop - "3-channel" is flawed on several levels

the whole point is to control the difference V across the driver terminals and the "3-channel" gnd buffer is a error source in series with the drivers
you really, really want the R,L channel signal gnd reference, feedback gnd and driver return to be the same electrical "point"

star.jpg


I'm afraid that "switching" between different philosophies for ground and driver return current is unlikely to be easy - there is more than one wire that needs to be switched - and the very fact of routing the wires to a switch will compromise one (or both) schemes
 
Last edited:
BIG THANKS TO DRM1962!!!
This fine gentleman gifted me the CRDs that I needed for my power supply and I'm forever grateful for his generosity!

I got my sigma22 pcb in the mail and have been slinging some solder for the past couple hours.



Also, I bought another case...






... and this one has vents!!


It also has sweet hinging actions and it's longer; I've been playing with the idea of mounting the toroid on-edge at the back of the case. Downside is that this box is of welded construction (the other one is held by screws) so I'll have to make some sort of sled that I can build on, then slide everything into the case.

The on switch (toggle type) is located on the right side, inside the door! I guess this bit of equipment was something you turned on, calibrated, and forgot about.

Maybe I'll use the first box for a tube pre-amp :scratch2:
 
Disabled Account
Joined 2006
Ref: The σ22 Regulated Power Supply

Regarding changing the values, most folks seem to think that just adding caps does very little, and series resistors should be added with the caps. There used to be some good info on these 'snubbers' floating around the 'net. (I think it was from Jim Hagerman of Hagtech).

Alternatively if you use Schottky diodes you should be able to delete the capacitors altogether.
 
Ref: The σ22 Regulated Power Supply

Regarding changing the values, most folks seem to think that just adding caps does very little, and series resistors should be added with the caps. There used to be some good info on these 'snubbers' floating around the 'net. (I think it was from Jim Hagerman of Hagtech).

Alternatively if you use Schottky diodes you should be able to delete the capacitors altogether.

Thanks Gordy; I can't seem to find much info on this, unfortunately--does anyone else care to chime in on this subject?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.