PCB for Aleph P 1.7 preamplifier

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.

I founded the pcbs for the Aleph P 1.7 preamplifier at http://web.vip.hr/pcb-design.vip , and I would like to build it.

The design is really awesome and it looks unbeliveable,
but like you see there is no power supply on the pcb.
I talked to Kristijan (guy who designed pcbs), and he said that with separate power supply (in another enclosure) preamplifier would sound even better, if that's possible.

Now I am little confused, I tought that all wires should be very short, and with separate enclosure they will be longer than in the same enclosure - with preamp.

Can anyone tell me which way should be better, and why ?
Is it worth it to install power supply in separate enclosure ?

Using short wires for the signal is a good idea. For the power supply, it's a lot less critical. The power supply puts out DC, which is not particularly influenced by long runs of cable.
Putting the power supply in a separate location helps by reducing hum (picked up from the transformer), and various kinds of high frequency noise (emitted by the diodes in the rectifier, etc.).
It can be a nuisance when you want to move things around, though. Consider having a detachable cable between the power supply and the circuit itself.

Power Supply wiring

With the greatest respect, I beg to differ. The regulated supply and or filter caps should be very close to the circuit with short wires for low impedance and good regulation and stability. Isolation of radiated transformer noise is a good idea but other factors must also be considered. The line rejection is made worse by signal currents in the power supply creating voltages in the wire's series impedance. RF noise from the supply is not reduced since it is conducted by the wire. If this wire is not shielded more RF can be picked up by the cable. Power supply design is as important as "signal path" to how a circuit sounds. You do a great diservice simplifing this subject as can have great impact on the performance of a circuit.

how to solve all problems

So how do you build a better power supply? What if I put some large caps in the amp enclosure across the power rails in addition to the power supply caps? and how do you prevent the rf from being transmitted through the wire to the amp? I'm talking about an unregulated supply, so there's no feedback path

Separate power supply

I am quite surprised that from all audio components only preamplifiers have separate power supplies (in general). This was maybe important when RIAA stages were part of the preamp, but now they are usually not and if someone needs phono preamp it's another separate box with it's own supply.
SS preamp circuits are not much different than power amps front end circuits. Yet those front end circuits are never separate from the amp (the only exeption I'm aware of is Coda amp) and they work quite happily in spite of the big toroid transformer inside the amp.
My impression is that separate PS for the preamp has only sense if there is not enough space inside preamp enclosure or the manufacturer wants to elevate the price of the product.
I have never seen separate power supply for CD transport or DAC (again the only exeption I know is Lab47) although those units are much more sensitive to transformer noise than simple preamp circuit.
I can say from my experience that power supply has very big influence on the sound (probably the same as the circuit itself). The same with wires. My suggestion is that it's better to do it in one box. But if someone wants to do it separately good idea is to put transformer, rect. bridge and some caps in one box and then the rest of the caps and active regulation circuit in the same box as the preamp. Also I recomment using split bobins transformers (not toroidals) and separate supplies for right and left channels.
one counterexample...

one example of amps, phono stages, & cdp's that use outboard power supplies is naim audio. In fact I think the only thing they make with a ps included is the integrated amp.

but I guess I'm hearing that you don't gain much by moving the ps to a separate box, and you could hurt some if done wrong.
You just confirmed what I said. You could give me only one example. To tell the truth Naim equipment never caught my eye. It doesn't seem to be popular were I live.
If you shield supply of your Aleph preamp inside chassis I don't think it would be different than separate supply.
Although those boards look pretty big and you might not have enough space. ;)
Sorry if I was misleading above. I'm at work and have to fit my replies into spare scraps of time between doing other things.
My usual strategy is to have the transformer, rectifier, and bulk caps in one box, then regulation and more caps in the same chassis as the circuit.
Another nice trick is to use circuit topologies that inherently cancel out signal junk on the rails. A good example of this is a differential. The signal passed though the load resistors--one positive and one negative--cancel, leaving only a net DC drain on the power supply. If you can contrive an entire circuit that has only a DC draw, you've got a lot more lattitude as to what to do with the power supply, although I still tend to regulate things.
Note that a lot of regulator circuits aren't stable without a cap nearby and plan your layout accordingly.
Amps rarely have remote power supplies because of the current draw involved. It causes too much voltage drop through the umbilical. Smallish solid state amps are about all you ever see with remote power supplies. Class A and large amps are a whole 'nother ball of wax. Tube amps are a better candidate for this because they use more voltage and less current.

Good Save Grey!

I agree with everything except I think that seperate supplies for amps are fine with the game plan you outlined. Even line level and power amps can pick up noise from the magnetic field from toroid transformers and AC wiring. My Aleph 3 measures more noise in one channel than the other due to closer proximity of one channels circuitry to the AC wiring than the other. Jocko once demonstrated a dramactic decrease in noise and improvement in sound on a Threshold amp by moving the unbolted toroid power transformer about 4 inches further from the amplifier circuitry. It suprised both of us how easy this was to measure and hear. There are also advantage in decoupling the transformer vibrations from the rest of the amp.


You said "also I recomment using split bobins transformers (not toroidals) and separate supplies for right and left channels".

Do you mean on the E-I type transformer when you mentioned split bobins transformer ?

Why they should be better for the preamplifiers ?

Also, I heard that E-I type transformers are more critical to place near preamp circuit and that they can cause more buzzing than toroidals.
Is this true ?
Kristijan's pcb is a kind of big, so the transformer would be close to it if all will be placed into same box.

If power supply would be placed into separate enclosure, what type of wires should be used (regular or shield type) for power connection ?

To answer your question about transformers I will quote interview with John Curl, a designer behind Mark Levinson JC-2 and Vendetta preamplifier:
"When we talk about transformers we have to separate power amps and preamps. While they are not perfect, Toroidal transformers are logical choice for power amplifiers because they are very efficient, they tend to have a fairly low hum field, and they're readily available in large power ratings.
For preamplifiers and other line-level components, the old type EI transformers or what's called a D-core or split C-core transformer is actually better than a toroid. First of all, they tend to be more compact, and second, and perhaps more importantly, they have very low capacitance between the windings.
This can be a problem when Toroids are used in low level signal applications; the windings are on top of each other so they talk to each other. And today the AC power is so dirty. We fix that by using a transformer that isolates the winding, which is important on low level circuits. The old style EI or the new C- or D-core is the ultimate in that respect-as long as it is a dual-bobbin winding with physically separate bobbins for the primary and for secondary. It makes big difference in sound quality."
I'm only using split bobbin transformers for line level equipment now, never had problem with buzzing.
Retired diyAudio Moderator
Joined 2002
Speaking of toroids,

When I had my custom transformers made, I had the choice for a large diameter, low profile core, or a standard diameter, standard height core. I wonder if the dimensions of the transformer core make any difference in capacitance and other factors.

I picked up a custom low profile small toroid for my bosoz. The picture of it is at:


It is the one on top of my two Aleph 2 transformers.

diyAudio Editor
Joined 2001
Paid Member
Hey Brian:

Those are mighty fine looking transformers.

There was some discussion on another thread where someone
had ordered a 60 volt transformer and had trouble regulating it to the correct voltage. (it was a lot bigger than Nelson recommends, and put out more than it's rated voltage under such a small load) Did you order one total for both channels? Is it 60 volt each leg?
Please keep us informed about how it works. better hurry- schools almost out!

Retired diyAudio Moderator
Joined 2002
For the BOSOZ, I ordered the transformer so that it gives +/- 80v dc unregulated (circuit regulates it down to +/- 60v) and assembled it as shown in the article, and it powers up with no problems. I have not done extensive testing yet, and I will when I get back. All that I did was plug it to a function generator and tied the output to a scope, and appeared to work properly.

For my Aleph 2 project, I ordered two 1kVA transformers with 2 x 35v secondaries, which will give around 45v under load.

I am almost done with my chassis, and my boards are partially populated. I lost my IRF9610 devices, so I will have to buy more and match them. I put the project in storage, and I am now in France for 10 weeks (study abroad - Georgia Tech Lorraine). I will finish my Aleph when I get back.

My heatsinks are larger and much thicker and heavier then the ones that Wayne Sankey used for his.

Comparing cross section:

My heatsinks

Wayne Sankey's heatsinks

The dimensions of his are 7x9 and mine are 8x9. I am going to test a channel using two heatsinks, and if I can, I will put the two channels in one chassis. This will provide me with another chassis for 2 more channels, that way, later on I can bi-amp my speakers.

I believe that two channels could run on a 1kVA transformer for the Aleph 2. There would still be headroom, but not nearly as much. I actually might be able to fit two transformers into my one chassis (9 x 18 internal footprint), assuming I bought bigger caps.

I have 10 weeks to think about it while I take classes and explore Europe.

Hey Brian, if I remember correctly from another thread, you were going to use 4 heatsinks for each channel. You've got the blessing from the master himself. To use 2 heatsinks only would means 12 transistors on each heatsink - that's just too hot ..... just from imagining it!
Retired diyAudio Moderator
Joined 2002
My original plan is 4 heatsinks per channel. With 4 heatsinks per channel, there are 3 devices per heatsinks. With 2 heatsinks per channel, there are 6 devices per heatsink. If you sit down and think of it, 6 devices on one 8"x9" heatsink really is not very many.

I will assemble it with one channel using 2 heatsinks, and they test it, and if it runs too hot, I will just use all 4 for one channel.

From seeing other designs, the 4 heatsinks should be enough for both channels. I have two chassis with 4 heatsinks each almost completely assembled now.


(also now available at: bbell@georgiatech-metz.fr )
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.