advertisement


Starfish revival

jpk

pfm Member
As promised I tried to reconstruct the starfish circuit and open a new starfish thread. I used the guide-PDF and BOM v1.2, but I don't have full schematics of the starfish, and I am not a circuit designer, so there might be mistakes in the following diagrams. First the audio circuit, a standard Naim style 729 (TA, time aligned buffer) and 321 (gain stage) powered symmetrically from multiple sources:

6HHJVwV.jpg


I am not sure if all these these parts should return to the signal ground or if it's better to pull some of them to the noisier power ground (and connect them at one point together). The starfish BOM says C12 and C14 shall not be populated but shorted with wire links: what happens in that case to R20 and R23? BTW I used "starfish minus 100" as part references in this diagram (i.e. C12 is C112 on the starfish schematic).

I ditched the Traco DC-DC which was originally part of the starfish and instead start with a simple hicap style PSU with 2 separate rails to feed the negative and posive supplies of the audio circuit:

Io5J3CT.jpg


The original starfish used 1086s as local regulators, I draw them for negativ und positive rails:

B5KCwiN.jpg


Or are 1763s better...? The solder jumpers are there in case fixed voltage regulators are used:

2PMuuKD.jpg


The grounds can be routed to one star point in the preamp case if PSUs are done this way. A simpler solution would be to omit the local regulators and use only one stage of regulation in the PSU case:

HLEwpOv.jpg


The audio circuits needs 4 positive and 3 negative supplies per channel, so one would have to duplicate the shown regulator circuits. Or use advanced solutions such as TeddyRegs, SuperTeddyRegs or ALWSRs etc.

Questions:
- where can I order good quality PCBs from gerber files for audio circuits? I work on kicad and live in Europe.
- if replacing the volume pot with a stepped attenuator: could the DC coupling be reduced as suggested in this thread?
- should we consider to replace the TA stage with something like a B4 or a Kuartlotron?
- should we consider to just ditch the TA stage as shown in the air guitariste schematic?
- what about the current source mod for the gain stage as suggested in this post?
- what about the current source mod for the TA stage as suggested in this thread?

[questions edited]
 
Last edited:
Seedstudio for pcbs. I have a populated starfish board lying round if you'd like it as reference.
 
Notice that 'Kuartlotron' uses the BC560 like the B4...maybe we do need to get a last time buy of them!
 
My starfish uses two super teddy regs to feed the local regs which sounds a lot better than the traco.
I use s goldpoint stepped attenuator instead of the volume pot.

Pete
 
My starfish uses two super teddy regs to feed the local regs which sounds a lot better than the traco.
I use s goldpoint stepped attenuator instead of the volume pot.

Pete

I went a bit mad with mine and used a load of Teddy regs instead of the standard LM3XX local regs. It's long gone now though :(
 
I went a bit mad with mine and used a load of Teddy regs instead of the standard LM3XX local regs. It's long gone now though :(

What happened to yours? Did you just move on to something else?

I have a Neil Jadman Mini-Board, B4 and a StarFish which I hope to get round to one day.....how do they compare?
 
What happened to yours? Did you just move on to something else?

I have a Neil Jadman Mini-Board, B4 and a StarFish which I hope to get round to one day.....how do they compare?

Girlfriend kept nagging me about all the ugly black boxes, so I sold it all :(

I'm now more than happy with my modified Quad 306 and passive preamp.
 
Some renderings, first the (mono) 729/321 board:

d5Chjmc.jpg


Parts mostly original, but changed inputs to 2u2 film caps (shown as red rect type) and power rail caps to 10u tants. The PSU board has 2 separate supplies (good for one channel):

QdkJFmX.jpg


The regulator board (seven 1086s again for one channel):

JZLfnw0.jpg
 
It's great that you left the regs out of design so the main board has small footprint. I think that with such small size we can even call them 'local' regulators.
One of the features of the original Starfish is the option for local regulation for every part of the circuit (IIRC 14 regs all together). I would choose the best that can fit. Gyrator plus LM317/337 is the choice for B4 too but remember B4 has very good PSRR and I don't think that's the case with Starfish.
 
Hi Jun,

I'm still digesting some of this, but I have a couple of observations.

I can't see any value in separate regulators for the negative rail. It supplies 5 CCSs (3 double-BJT types and 2 resistive types), so there's heaps of PSRR built in there.

The positive rail is another story.

On dropping the 729/TA buffer, I think we want the filtering, but probably don't need the buffering (who runs a tape loop these days?). I've not yet wrapped my head around whether or not the Air Guitariste is providing the filtering without the buffering, or is dropping both.

The current source mod for the gain stage is replacing a tight-feedback (high impedance) constant current source with a lower impedance but voltage-controlled current source. I haven't read up on it yet, but that's going to throw away some PSRR, no? (You could recover that by feeding the LED with a CCS -- an E-452 between R9 and the LED would do nicely.) But either way it's starting to become a different preamp, and if I'm not re-creating an icon then I'd rather just use my HPA-1 clone. ;)

Doing the 729 CCSs would appear to have even less value.

Anyone know what the story with C7/C8 is? Is that just because a 120pF capacitor is hard to find?

Cheers,
Jeff.
 
I read up more on Jim's current source mod and I see using a CCS to feed the LED is discussed in the thread -- but Jim felt the improvement even without it was significant.
 
can't see any value in separate regulators for the negative rail.

The starfish cloned the NAC552 in this regard I thought...? But here is a pic of the 552, it has golden Caps for feeback and power rail decoupling

NAC552.jpg


So if I assume away the tape buffer and the feedback cap it seems there are 5 power sources per channel, that could make 4 positive and 1 negative rail...?
 
I don't think we can tell from that. You could have local decoupling caps even if the sources weren't separately regulated.

Do you have a picture of the insides of the 552's PSU box?
 
No, but the stereophile 2003 review says: "The NAC 552PS power supply, built specifically for the NAC 552, has a noise-reducing "split-rail" (ie, separate positive and negative supplies) power supply based on an 800VA toroidal power transformer "big enough to supply a 300 or 400W amplifier," according to Naim. Three large capacitors and 16 regulators supply power to the NAC 552's various sections. The split-rail design is said to keep signal levels clean, quiet, and thermally stable."
 
800VA? OK, so the PSU is spec'ed for marketing, not sound. Which makes all the other features suspect as well. :cool:

(But 16 regulators certainly sounds like separate ones for all the negative CCSs as well as the positive ones.)
 
800VA? OK, so the PSU is spec'ed for marketing, not sound. Which makes all the other features suspect as well. :cool:

:D

I owned the 552 long time ago. It sounded sooooo much better than the NAC52, I will never forget that difference... Since I can't buy it again I would love to diy something comparable ;)
 
I planned 14 regulators (7 per channel): could make 16 of them so the 2 remaining ones can be used for front panel leds, motorized pots, remote etc. (isn't the 800VA needed to lit up the green logo...????)

BTW my preamp will also need an input selector (2 sources are enough for me) and a remote controlled volume - probably with stepped resistors...?
 


advertisement


Back
Top