DIY Pass Labs Aleph Ono

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I just downloaded the service manual for the Pass Labs Aleph Ono. I'm very surprised I haven't seen more interest in this design, as I know there was just a thread asking about phono pre-amp circuits??

Anyway, I have an MM cartridge (Grado) and an amplifier with SE only inputs (Mark Levinson).. so I would not need either the moving coil section or the unity gain invertor portion of the circuit. That would leave me with the highly simple power supply and MM gain stage, which I could probably PCB and build in less than a week.

I would probably build a power supply chassis, a left chassis, and a right chassis, all small, and use the circuit exactly. Anyone thought of or done anything like this?

The unit was supposed to sound pretty darn good.

Thanks.. --arnach

So I redrew the schematic and PCB removing all the components I didn't need and it looks very nice and compact and will fit into a couple nice Thailand Audio ( chassis, went and got a price qoute for them.

I was surprised to see that this preamp dosen't have a dual supply, it is a single toroidal transformer (ok), but rectification and regulation (bad?). Is there any situation where this is good? Does this help channel matching?

In my experience and my feeling is to use completely seperate supplies including seperate transformers and make them identical as possible.

All in all with seperate supplies and even three chassis (two p/s, one audio) this will a very inexpensive project.

-- Aaron Finley
I had looked at the Ono (so where's Yoko?), but decided against it, as I wanted to pursue the balanced tube circuit I've been working on.
I'm all in favor of separate power supplies and regulation. For no more current than you're going to need for a stripped down version of the circuit, it shouldn't be expensive; I suspect that your problem will be finding a transformer *small* enough, current-wise, so that you're not wasting money. It might be better just to use one transformer, then have separate left & right regulators.
I, too, have heard good things about the sound of the circuit. It should be a worthy project.


Just got back from vacation and am going to start drawing the PCB tommorow. Right now I am collecting the various datasheets for the active components and having a rather tough time.

In so much as there are 27 resistors in my take of the design, and 8 of them I have deemed of special importance, I plan to make the 19 not-so-special resistors of the Holco variety, and the 8 of the Vishay variety.

Due to the low number of capacitors also I plan to use high-end capacitors throughout and very expensive high-high tolerance (haven't looked at the catalog yet, can you tell?) caps for the RIAA eq and roll-off.

As for transformers, no single transformer I can buy will not be a total waste of money so I plan to use two per channel. I am going to use the Talema encapsulated devices, they look nice and a relatively inexpensive.

-- Aaron
Hi Aaron - these guys (in the Czech Republic) seem to have all 3 Xono semis - 2SK389, 2SC1844, 2SA991. Don't know about data sheets but if they can supply the real thing does that matter?
Has anyone else sourced these semis?

I also found datasheet for 2SK389 - - no they don't have the others. 2SK389 appears to be a power device!!

The best I did with the others is:
2SA991 SI-P 60V 0.1A 0.5W 90MHz
2SC1844 SI-N 60V 0.1A 0.5W 100MHz

both from

So has anyone managed to substitute any of these semis from some place less esoteric - e.g. Farnell/Maplin etc?

I'm not sure on how closely these devices need to be matched. I think only Q5 is critical - and that's looked after by the dual package. Can anyone comment on this? (MM use only - I saw the bit on matching the MC stage semis)

BTW I found these using

Many thanks


[Edited by peted on 09-18-2001 at 05:25 PM]
NEC no longer list the 2SA991/2SC1844.
NEC Datasheets <a href=""><b>here</b></a>
They don't have a discontinued item listing either.

My data book list the following in addition to <b>peted</b> posted data
2SA991 NEC 400h<sub>FE</sub> | 5.5pF C<sub>ob</sub> | ECB pinout

2SC1844 NEC 400h<sub>FE</sub> | 4.8pF C<sub>ob</sub> | ECB pinout
Pinout is face on leads pointing down

My Japanese equivelents book lists
2SA991 = 2SA1016K (Sanyo), 2SA1083 (Hitachi) <2N5087(SAMS Listing)>
2SC1844 = 2SC2362 (Sanyo), 2SC2545 (Hitchi), 2SC2602 (Mitsubishi) <MPSA18 (SAMS listing )>

Hope this helps



Almost done specifiying parts.


So far, using Caddock TF020Rs for 8 resistor values, and Vishay S102s for 9 or 10 resistor values, the rest will be Holcos.

Basically the entire MC stage resistors and the RIAA stage is Caddock/Vishay.


RIAA caps will be 2% Soshin Silvered Mica.

.047uf bypass caps will be Multicap RTX polystryene/tin foil.

Misc 10pf, 100pf, 330pf caps will be 2% Soshin Silvered Mica.

Power supply bypass caps: 220uf will be standard Nichion Muses, and in important areas Black Gate Standards.

Main power supply caps, 1k, 3kuf, etc. will be Mallory capacitor grade.

What do you guys think?

-- Aaron Finley

Take that up with Thorsten. He swears by Silver Micas. Also they are in the tolerance I need.

But yes, I would like to use the RTX caps in the phono stage or even the TFTs. But they are in 10% tolerance..

Now I could either just use the 10% tolerance parts and forget about it, or I could order 1% parts but I only need about 6 and I would have to order about 30 minimum order (10 ea.). Can't afford it.

-- Aaron
There is a way to use 10% caps with good results. Just order matched pairs ( some dealers charge $1 extra for that). Then measure the capacitance and recalculate the resistance values accordingly. That way you could acheve tolerance in the range of .5% or better for parts.
I have built Aleph Ono two years ago. The active circuitry fitted on a board 1.65" by 2.75". No MC stage. no balance out. You probably know that you don't have to use coupling caps in the output. Just place trimpot in parallel with R4 in colector of Q1. You can adjust the offset pretty well and it's stable.
In the RIAA eq and roll off I used MIT RTXs, Caddocs and installed them on the separate board which also holds the active circuitry board and power supply caps. That way I can play with different caps and sizes without being limited by space. The layout for the active circuit is much simpler that way too. Came out well and the sound is good.
I've been also thinking about combining Ono with Pearl by using Pearl input stage and passive RIAA eq and using Ono output stage which looks more sofisticated.
As to silver micas I wouldn't use them without comparing to good polystyrene caps. My personal favourites are RTX and Hovlands.

I'm struggiling here.. I already have the resistors.

Well, I can use 1% Auricaps in all the right values.

Or I can use 5% Hovland Musicaps and a 1% Auricap for the .0027 value.

Or I can use MIT 10%s and cross my fingers.


-- Aaron Finley

P.S. Thanks for the info on your project. I'm using different types of Teflon substrates for the MC stage and MM stage and Arlon 25N for the power supply after extensive "research."
In stead of refactoring the Ono you could also
consider to build the pearl from Passdiy. I haven't studied the specs (signal to noise ratio etc.) and compared them, but the pearl uses the the 2sk170BL-fets for the first input stage, which have been selected for it's extreme low noise figure.

It's also safer, as it's already been tested while a dressed down Ono has not.

On the other hand, if you consider to use an MC element in the future it will be easier to have the ono already partially built. And maybe tweeking the Ono is more fun.
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.