Aleph-X builder's thread.

jh6you,
If you want to experiment with rail voltage vs. bias current, feel free. I've got more freedom in that than most--at least for the Aleph 2s--since they are water cooled and can dissipate a tremendous amount of heat. Just keep an eye on device dissipation; don't so wrapped up in the output that you forget the front end. It may need a little fine tuning as far as resistor values if you change the rail too much.
I've varied the bias quite a bit. If you lean it out, it starts sounding "hi-fi" which may suit some, but it's not to my taste. I've only varied the rails by about 5-8% or so. Not much to tell in that tight a range (by comparison, I've run the bias probably +-50%).
bertram,
The front end current source will be happy with any reasonable source of 9V. You could, for instance, use a voltage divider from one rail to ground or to the opposite rail--might be a good idea to have regulated rails. Zeners work. I haven't used them in this application, but a voltage reference should be fine.
I used the blue Panasonic metal film resistors from Digikey. What might be available where you are, I do not know.
The MPSA18 is not particularly critical. Nelson uses the Zetex 450 in some of the Aleph schematics. That would be perfect. The MPSA18 is a pretty ordinary 45V NPN. The circuit won't come unglued if you toss in another NPN. I used MPSA18s because I've got a zillion of them.
Two rectifier bridges would be great. Or you can use just one. Whatever suits you.
Wind your own chokes. It's much cheaper.
Although I'm sure Nelson wouldn't turn down a good French wine, I believe he's on record as leaning towards the California product. I'm prone to French, Australian, and Californian wines, but not necessarily in that order. Just depends on what's going on. Being a fan of red wines, myself, Germany doesn't get too high on the list....however, there are some stunning German beers.

Grey
 
This forum has become one of the most interesting places for DIY audio in the world. It is always a joy to see how people of different nations and cultures can participate in the exchange of ideas and experiences. It has helped making DIY audio a very dynamic and sophisticated affair. People with all kind of levels and skills have learned so much from the contributions of men like GRollins, Dieckmann and so many other members, that we must try to protect this forum from unwanted resentments.

When sharing ideas in a forum, one is bound to engage with people that have different backgrounds, methods, and characters than your own. If it becomes difficult to respect that, the best next thing is to ignore the other person. That at least prevents the forum from being obstructed by personal resentments, which really do not help any of us...

This very thread will become more and more important when so many members will start constructing their Aleph-X's in these coming weeks and months.
I hope that after the two days pause, both (really highly estimated) members will continue to contribute in a postive way to this thread. They have both (each in his own way) presented so much knowledge, experience, and inventions to this forum, that it would be a pity for all the other members when we would have to continue without them.

May this thread flourish just as much as the sound of all the newborn Aleph X's that will come to life!
 
The one and only
Joined 2001
Paid Member
Re: FET's to use

Jason Hubbard said:
Nelson said to not run below 25volt rails - was this in respect of 244's only (and therefore 044's are ok at say 19v)?

You can run them at any rail voltage you like, but below 25
volts or so you start into the part of the curve where every
volt counts, so I recommend that at least as a + to - figure.
 
I need a source for .22 ohm wirewounds for the X. I was able to get both .2 and .25 Vishays so which of these would be the preferable one to use in a stock amp?

Also as far as matching these low ohm resistors goes would a 5-1/2 digit meter suffice with direct resistance readings or is it better to measure voltage drop across them(with the same 5-1/2 digit meter of course......?

Thanks for the help!
Mark
 
I'm a little confused... It seems that the First Watt Aleph X's will
have 24 volt rails and run at aprrox. 15 watts.

15 Volt rails for the DIY Aleph X gets you close to 35 watts

these figures are just off the top of my head... but there seems
to be a discrepency between the two amps in this area.

I know these aren't identical amps, but why the difference?

thanks for putting up with questions from the back of the class ;)


m.
 
Mark,
Choose the resistor value according to how much bias you want and how much heat you can safely dissipate. Lower resistor values (all things being equal, meaning you don't fiddle the current source, etc.) will, of course, lead to higher bias current. Look at your speaker load impedance and decide what its minimum value is. Work backwards from there, then throw in a bit of extra bias, just for spice.
5 1/2 digits will do fine for matching. What tolerance are the Vishays? 1%? I wouldn't even bother matching them, myself, but then I'm short on time. Someone with more time on their hands might feel that tweaking the R values would be worthwhile.
moe29,
I think it's safe to say that the First Watt Aleph-X will turn out to have a slightly different topology than the Aleph-X that I designed. I have at least two possibilities in mind as to what Nelson's up to, but haven't had time to think either idea through yet.
Jason,
No need to go for '044s unless you want to. Use the '244s.

Grey
 
Heat Protection for Alephs

I'm posting this question here because I recently got hifiZen's boards and may build the X next. However, I also bought some regular Aleph boards, courtesy of Coulomb, and might do that first.

Here is the question: The Aleph manuals are contradictory on heat protection: both the 5 and the 2 manuals say in one passage that a 70 deg. C thermal protection switch is installed, yet say in another that a 75 deg C switch is installed. Yesterday (before I discovered the contractictory passages) I bought a bunch of 70 deg C switches from DigiKey, thinking no matter what Aleph I build--2,5, or X--its going to get hot, even with the R Theta heat sinks at 2 per channel. Did I screw up in buying the lower temp value? Am I obsessing again?......Dang! Where are my meds?

Larry Wright
Seattle area
 
Thanks Grey for the biasing tips. My resistirs are 1% tolerance but I think I'll still take a few minuites and read em all and see exactly what I've got here. I'll probably start off using the .25 ohm jobs. I suppose I can always adjust the operating current slightly from there anyway.

Larry,
I reccomend building a regular Aleph first so you can get used to matching fets adjusting bias and such. They are still just about the best thing going out there. From what I've read the AX sounds a bit brighter and packs a bit more slam than a same size Aleph. I'm about to find out....hopefully this weekend.......

Mark
 
Hi Mark: I may take your advice, just to get another success under my belt after the ZenV4. My previous effort, an A-75, is sitting on the shelf awaiting my return to fix it after screwing up the power leads to the output stages at the very last minute (what a dummy thing to do). I actually did quite a lot of Vgs matching of devices for that and got some very tight groupings of all three Q types, especially the IRF230 and IRF9230 output devices. Still have quite a few of those matched devices and have mulled over the idea of using them, like Peter D., in an X. As Nelson recently suggested, I gave them about 5 seconds of test time, but made sure they were all about room temperature first.

As elsewhere posted in this forum, the pot adjustments on the A 75 were a little twitchy too. When I go back to the project, I am going to use multiturn pots to make it easier. I think if I do the X I will go that route with it too. Those blue and white Bourns (or whatever) single turn jobs are for the birds. Seems like the X (like the A 75) is hard enough to set without dealing with a balky pot too.

thanks for your comments. I may take the easy way out on this next project. Don't know yet.

Larry Wright
Seattle area
 
Capacitors

Howdy Gents,

I am getting parts together to start my Aleph-X in parallel with finishing my Zen IV. I would like to start by designing the power supply. I will design a board for the rectifiers and capacitors. I would like to use a good quality snap-in capacitor. The rails will be +15V and -15V and I would like to do a CLC filter with 47,000uF cap, 3 mH inductor and 47,000uF. Could anyone recommend a capacitor manufacturer? I would like to use better components. Would you recommend 105 degree caps? Thank you in advance.
 
Today I ordered a couple of dual secondary 16V transformers rated at 17.5A each secondary (they were only $45.00 each). Was it a sound purchase? I would like to use four devices per board. The AX spreadsheet tells me I can only get about 29W in 8Ohm with these and 15W in 4 Ohm. I would like to bias a bit higher but with lower supply voltage. Can I do something to the power supply to drop the voltage a bit? The spreadsheet suggests that the actual voltage at the Fets is a bit higher than the secondary. Maybe I should go with eight devices per channel. What's the concensus?
Sorry if the questions are a bit basic. This is my first amp. I did a few DIY speakers before but this is a new territory.
 
Re: Resistors and Capacitors

Mark A. Gulbrandsen said:
Thanks Grey for the biasing tips. My resistirs are 1% tolerance but I think I'll still take a few minuites and read em all and see exactly what I've got here. I'll probably start off using the .25 ohm jobs. I suppose I can always adjust the operating current slightly from there anyway.
Mark

Hi Mark,
As Grey said, with 1% tolerance, you'll be fine, even without measuring them. To be overkill, I'd chose resistors of twice the original value in 1% tolerance, and paralell 2 of them (or 3 with three times the value, etc... - I used 4 //ed ;) ). Two advantages here IMHO : a larger saftey margin re. dissipation and stability, and the couple of resistors will average the total resistance value in all the places they will be used, leading to a better match between resistances values all over the amp. I felt better this way, with all the time I spent closely matching the mosfets :)

kilowattski said:
Could anyone recommend a capacitor manufacturer? I would like to use better components. Would you recommend 105 degree caps? Thank you in advance.

Hi Guy,
I "usually" use Panasonic TSHA snapin caps. 105°C rated (mandatory in this use IMHO, longer life), can be easily sourced (Digikey), and very robust. Try to bypass each cap with a small film cap (100nF/MKF), and you're in ;)

Hope this helps
 
Hey Mark, I like you three phase idea. In my new house, my listening room backs up to the attic, and I was considering an auto parts AX. I could use the radiator and electric fan from a front wheel drive car and I even thought about using a electric motor to spin a car altinator instead of huge transformers. You can vary the output voltage by the feild voltage and the output is three phase if you connect straight to the stator and make your own bridge. It may take some surgery to convert from delta to wye. Even without capacitors, rectified three phase from a car alt only has like 15% ripple and would be much higer frequency (like 2000 hz) meaning much less needed capacitance.
Hummmm. This idea needs more developement. I could build an ugly as $hit frankenamp and have it out of sight on the other side of the wall in the atic (I have a 2' x 6'8" door right next to the amps home) and it wouldn't offend the wifes sence of decor.