Sec. Transformer Voltage for Aleph 5

[Zapped]

Saw a great deal the other day for a transformer rated at 25vac at 625va on the secondary side. I was wondering whether this would present a problem for an Aleph 5, which appears to require a 24vac secondary to get to the 34 volt rail cited in the Technical Manual. Is this difference significant or insignificant? Would use of an inductor in the filtering be advised (and wouldn't that drop the rail level a little?)? Or should one just go for the 35+volt rail that results from the 25vac transformer and rejoice over the great deal? Naturally, I'd use caps rated at 50 volts or so, for adequate margin.

(I got a good deal on the sinks from Peter D., great price on the caps from B.G. Micro, and cheap aluminum plate from Boeing Surplus. It would be nice to get cheap trafos to stay with this pattern!!)

Larry Wright
Seattle area

 

[Leeuwarden]

I'm using 2 x 30V, 625 VA. Works great on my Aleph 5. So I guess 25V is no problem.

Jan

 

[Holger Barske]

The usual hints here go for 28V secondaries. But there is some room for variations of course.

 

[Zapped]

That is welcome news. Thank you for your response. I think I'll go buy those 25vac budget-saving transformers while they're still available.

Larry Wright
Seattle area

 

[roddyama]

The usual hints here go for 28V secondaries. But there is some room for variations of course.

I have 2 - 500VA 28-0-28 for my Aleph5 monoblocks.

 

[Zapped]

Thanks Jan, Holger, and Rodd. I guess this circuitry must be noncritical enough that there is "wiggle room" on the spec's?

By the way, I thought I understood how to estimate the rail voltage in this case, by multiplying the secondary AC voltage by 1.414. (For example, with the ZenV4 the spec was dual 18vac secondaries, yielding roughly 25vdc each (theoretical) after rectification and filtering, times two, for about 50vdc raw supply and ready for regulation down to 42. Right? I think I got 40 actual, but that's pretty close.) But in this Aleph 5 case, the 28vac mentioned times 1.414 yields a theoretical 39.6vdc, quite a bit above the 34vdc rails in the Aleph 5 schematic. A 30vac secondary by this thinking would result in about 42 vdc rails. What is wrong with my estimation method? (other than an unreasonable reliance on theory and not empirical results). I saw a Rod Elliott article in which he said the real multiple should be 1.1, rather than 1.414, but that doesn't seem to make sense in the Zen case, nor agree with a table I have from Toroid Corp of Maryland, specificying "typical" dc values for their designs after bridge and cap filtering. And 28vac times 1.1 would be 30.8 volts.
So much for that. ??

If the bias for this design requires "no adjustment" and the operation is set by "physical constants" (not sure what this is here; fixed resistor? diode junction characteristics?), does that mean that a rail voltage variation of 4 or 5 volts will not affect the bias level, or would I need to make circuit value changes?

Confused AND Sleepless in Seattle
Larry Wright

 

[Holger Barske]

Following NP's (and others) statements here, the supply rail voltage will be transformer voltage multiplied by 1.2 to 1.3, depending on some variables like transformer quality, rectifier losses and, if you go for it, voltage drop across the choke inductor.
In my experience, this is pretty close to reality.

 

[Zapped]

OK. This is the rule of thumb for calculations then, eh? I see where the 28vac fits in nicely. Since I can get a 25vac or a 30vac transformer at a good price from Parts Express, and both depart from the practical ideal, I wonder which is better, a couple of volts too low or a couple too high. Lemme see, where did I put that 50 cent coin....

thanks for the clarification !

Larry Wright
Seattle area

 

[wuffwaff]

zapped,

if your speakers are above 7-8 ohms go for 30V, if they are more like 4 ohms go for 25V. This will give you the chance to get higher power where you need it without changing total dissipation.

william

 

[Zapped]

My speakers are Magneplanars, 4 ohms purely resistive. Got another pair of really old ones upstairs that might figure in someday; and they are 6 ohms. Sounds like 25vac is the optimum, then, no matter which pair I drive.

Many thanks for the help in puzzling through this. I even looked in my copy of Art of Electronics for some guidance like this, but came up empty. (Parts of that book are pretty impenetrable for me anyway. Its good as a doorstop, or for curls or step-ups, however.....just kidding.)

Larry Wright
Seattle area

 

[BDP]

I built my Aleph with 25VAC secondaries and 3 amp bias. The DC voltage came out to be 33 volts. The capacitors are 33,000mF with 0.56mH inductors in between (surplus, very cheap and large guage). The amount of AC ripple at the first capacitor is 153 mV. The second capacitor after the inductor sees 16.6 mv. Thats roughly a -20 dB reducton in ripple. If you do the math for XC=1/2pi F C and XL=2 pi F L, frequency being 120 Hz, it works out to be correct. The inductors XC has about 10 times the reactance as the capacitor so that gives you about -20dB. The amps are very quiet.

Zapped, I'm in the Vancouver, Washington area. If I can help e-mail me.

BDP