PCB ground planes

[Cqwet Dbdfte]

Passive RIAA, sounds good. No feedback, fixed bias via LEDs. Tube loads are taken from "top" of current sampling resistors, avoiding load currents in the sampling.
Input triode connected pentodes running at 20mA, their gain is dependent on selection typ 85, some as high as 93. With low-ish plate supply I still can can get 11 V p-p, so any clipping would be in the next stage, gain of 33. The -3dB point of input stage is around 140KHz.
The 1KHz gain about 50dB. MC transfomers add another 27dB.
The beast pulls 23W out of the wall :-o
The current regulators are using LM431 types. Their gain varies widely between mfg's. The LT's part has the highest gain by a fat margin. Output noise is sub 10 Hz dominated, and around 1.5 mV unattenuated.

 

[Mr. Moscode]

The cathodes are on top of the LEDS? Bypassed with a cap? B+v?
What's a current sampling resistor?
Could share the schematic?
Just curious. I've built my share of RIAA phono preamps. Mine did 72 db gain at 1KHZ if needed. 3 tubes and a few transistors. No step up xfmer. Look up the Super It and Minuet in A.
G

 

[Cqwet Dbdfte]

I use a 431 controlling a HV MOSFET current per mfg's data sheet.
The voltage across the 431 limited by a Zener, an LND150 feeds the 431 and Zener, not a resistor.
B+ 190V. No bypassing on cathode LEDs, may try some high value OSCON, but LED impedance is very low, so that may not do much.
SMPS has a output series 300 w.w. resistor then 440uF, along with several 4uH ferrite inductors and nF caps to tame the HF from it.
Constant current loads are not that sensitive to PSU noise.

 

[Mr. Moscode]

Led Impedance is low but I've found the sound is less edgy bypassed, or in my case I just use a resistor and 330 uf cap bypassed with a film for max gain. I tried the led but didn't like the sound.

For a power draw that low a simple analogue PS is adequate. I don't see an advantage for a SMPS. A linear is sooo much quieter. Use a regulator into a secondary op amp regulator for uber quiet B+.

You're right about cc loads. They have hi noise rejection. But still . . . .
If you're happy with the result the more power to you. Good job in any case.

 

[Cqwet Dbdfte]

I'm still in experimental mode, the PSU choice was just a challenge, and it is VERY small, a 60Hz job would be toroidal, R - core and remoted to another box, as H fields are bitch to shield.
I yet have to measure intermods, I need more objective data. A low current negative hi Z grid voltage reg may be simpler to filter out than something drawing 20mA. No cap is perfect, nano amps is easier to clean than milliamps.
I have no particular affinity for classic tube circuits, only the tubes, old school had a lot of limitations in part choices. No reason to repeat those now imaginary limitations.
Kudos to Pete Millet for his deep dive into RF pentodes.
.

 

[Mr. Moscode]

When you're dealing with uber gain outboarding the psu is the way to go.
You can get a boat load of gain out of a pentode.
G

 

[Cqwet Dbdfte]

Look at Pete Millets list of pentodes, same gain as in triode, he also lists harmonic plots. CC loads extracts all the gain a tube has. I did this MOSFET+431+LND150 thing over 30 years ago when I wanted a big swing feeding 211 push-pull triodes. Works for low noise too.
I have a stash of old LM611, OP + a separate reference, where you can clean up the reference real good, and the OP can feed the MOS gate.
Same type part could be used for negative reg too.
Thanks for your comments
.

 

[Mr. Moscode]

I did this MOSFET+431+LND150 thing over 30 years ago when I wanted a big swing feeding 211 push-pull triodes.

I'm not familiar with that "thing". Could you share a sketch so I can see the topology? Sounds interesting.

 

[Mr. Moscode]

BTW- a pentode is , for all intents and purposes, a 2 triode cascode. Just with 1 cathode. Works the same way, curves the same way, too.

 

[Cqwet Dbdfte]

Yep, got those in my quad 6550 amps, thanks to Joe Curcio's design.

 

[Mr. Moscode]

I have that schem somewhere. I think that was in a Dyna Stereo 70. Sold it. Someone's happy with it.
I'm not a xfmer guy. Best amps I ever heard and worked on are the Futterman OTLs circuit. With my redesign I got 200 watts triode out of 10 EL 509s.

Happy Holidaze to everyone.
G

 

[Cqwet Dbdfte]

Output transformers are expensive, big, complicated, limited freq range and power, but provides voltage isolation, high freq filter, and load matching to typical speakers. Seen examples of direct driving electrostatic speakers with crazy HV tubes.
If voltage isolation is not an objective a transmission line type transformer could be used for very wide bandwidth. Designing and constructing these would not be a lightweight task.
With a good subwoofer the power demands drops substantially and the ST-70 type power should be enough, if sane speakers are used, not the 82dB/1W/m types with a 10 lbs. crossovers, in a domestic setting.

 

[abbey road d enfer]

a transmission line type transformer could be used

Do you have an example of such a circuit?

 

[Cqwet Dbdfte]

I read about it for RF some thirty years ago and started a design for LF, I have notes somewhere, I never proceeded to produce it. It makes a nice impedance transformation, but misses out on DC isolation, requiring some DC servo'd power supplies. Perhaps worth the effort, but I took another route.

 

[abbey road d enfer]

As far as I know, transmission-line based impedance conversion works only within a restricted frequency range. May be possible for tweeters, but not much else.

 

[Cqwet Dbdfte]

Although cumbersome, it could be made with the same size "iron" as a regular transformer. The same principles applies. But taking off a decade of bandwidth off the low end of the transformer would ease the design, most of the core is for low freq, where a class-D sub driver would work fine flapping the curtains, and not get mixed in with midrange.
Flat power bandwidth makes sense until you look at program content, most energy is expended in the low end. 200W at 20KHz is a nightmare for the dolphins and the bats.

 

[abbey road d enfer]

I think we're talking about different animals. A transmission-line impedance converter is not a transformer. It actually does not rely on coupled inductors.

 

[Cqwet Dbdfte]

Definitionally every impedance conversion is a transformation, and yes cores are used.
https://www.highfrequencyelectronics.com/Feb04/HFE0204_Sevick.pdf

 

[FIX]

Star groun is a notion based on safety. It is definitely not the best solution for low-noise electronics, compared to hierarchical ground.
However, most tube circuits are low-current, even a 100W tube power amp operates with less than 500mA, except heaters. However heaters are a separate circuit, which is easy to make sure their current does not pollute the other parts. With such low currents, a rather flimsy ground circuit would still offer good performance, but nothing justifies makinf star ground a preferred choice.


I believe the name "star ground" elicits grandeur, and allows its heralds to stand up, but really they should know it's good for safety, and poor for noise and other aspects of performance. Does one really want all their connections to ground going back and forth through lengths of wire?

I believe it's a manufacturing decision, that avoids creating large blobs of solder after wave soldering.

Maybe I misunderstand you. Does hatch mean "allowed to be crossed by other tracks"? In that case, that would be a wise decision, it it allows separating grounds that are not supposed to be mixed.

Don't overestimate the value of shielding. Indeed sensitive nodes need to be protected, but remember that the most important shielding is the one done by the case/chassis.
Not too long ago (1970's), ground planes were almost unknown in audio equipment. Many brilliant pieces of equipment had single-sided PCB's. Perhaps the most significant reason for adopting ground planes in audio circuits is because it reduces chemical waste.
Actually the Brits were conscious of that and adopted a style of layout that left most of the copper intact.

DC heaters do not radiate AC, so tehy can be routed almost any old way.

Yes because twisted pairs reduce magnetic radiation.

True. One big thing about ground plane, anywhere you have any kind of opening in the board (transformer) make sure the loop is cut, as it will pick up stray fields. This will induce hum into the grounds and you'll never get rid of it. Just think of every loop as one winding of a transformer. Just like the Weller soldering guns, one loop creates a lot of current and almost no voltage.

 

[Cqwet Dbdfte]

The same argument has been done for hashed ground planes, many local loops created, relevant at some frequency...
The idea behind "star" ground elimination of voltage differentials, where there should be none. The words"reference plane" and "common" are used to mark a difference to the obfuscated word "ground'" which has a few meanings.
My MC preamp PCB has besides top and bottom poured ground planes, with controlled continuity around obstructions, backed up by a multiple "ground" exposed vias, connecting fat aluminum spacers to a 2mm aluminum plate, thus unlikely to produce any significant ground potential differences.